Intensive Psychotherapy of The Borderline Patient

INTENSIVE
PSYCHOTHERAPY
of the BORDERLINE PATIENT
RICHARD D. CHESSICK, M.D.

Copyright 1977 Richard D. Chessick
e-Book 2017 International Psychotherapy Institute
All Rights Reserved
This e-book contains material protected under International and

Federal Copyright Laws and Treaties. This e-book is intended for
personal use only. Any unauthorized reprint or use of this
material is prohibited. No part of this book may be used in any
commercial manner without express permission of the author.
Scholarly use of quotations must have proper attribution to the
published work. This work may not be deconstructed, reverse
engineered or reproduced in any other format.
Created in the United States of America

THIS BOOK IS DEDICATED
TO MY MOTHER
www.freepsychotherapybooks.org 4
Table of Contents
Part I HISTORICAL AND CLINICAL DESCRIPTION

1 Introduction to the Concept
2 Etiology
3 State of the Ego
4 Typical Borderline Complaints
Part II METAPSYCHOLOGY
5 Basic Metapsychology
6 Metapsychology of Narcissism
7 Disputes and Disagreements
8 Developmental Failure in the Borderline Patient
Part III DEVELOPMENTAL PATHOLOGY
9 Early Ego Development and Projective Identification
10 Internalized Object Relations
11 Unresolved Metapsychological Problems: My Views
12 Clinical Material
Part IV PSYCHOTHERAPYGENERAL APPROACH

13 Overview of the Psychotherapy of the Borderline
Patient
14 Psychoanalytically Informed Psychotherapy
15 Therapeutic Regression
16 The Psychic Field of the Psychotherapist
Part V PSYCHOTHERAPYSPECIAL PROBLEMS

17 Transference in the Borderline Patient
18 Transference in the Narcissistic Personality Disorder
19 Ambience of the Treatment
20 Rage and Externalization
21 Countertransference
22 Helpful Clinical Suggestions
23 Improvement and Repair
Part VI METAPSYCHIATRY
24 Concepts of Cure in Intensive Psychotherapy
25 Philosophy of Science
26 Metapsychiatry and Beyond
References
We are still underestimating the pathogenicity, but
also the character-building, the personality-
integrative role of preverbal levels of development;
and we are underestimating in particular the
importance of ego and superego precursorsand
especially their capacity for creating hard-to-
decipher proclivities to intrapsychic conflicts!
Margaret Mahler
Symbiosis and Individuation

Part I
HISTORICAL AND CLINICAL
DESCRIPTION
CHAPTER 1
Introduction to the Concept
The concept of the borderline patient is poorly understood and

vague, and it is almost lost in a semantic morass these days. Some
psychiatrists wish to entirely discard the termborderline of what?

However, semantic disputes will not make the concept go away, and
the present book is dedicated to improvement of our understanding
and treatment of the borderline patient. It is written from the point of
view of the psychotherapist; the clinician who treats patients day in
and day out over the years and who wishes to understand the
application of various theoretical conceptions about the borderline
patient to the office practice of psychoanalytically informed
psychotherapy.
The best overall review article (Gunderson and Singer 1975),
which attempts to identify areas of agreement in the literature,
reviews eighty-seven referencesmany of which on careful

examination contradict each other. This overview paper enumerates a
number of features most authors believe seem to characterize
borderline patients. These features are
1. The presence of intense affect, usually of a strongly hostile or

depressed nature
2. A history of impulsive behavior of all sorts
3. Fairly good social adaptiveness (at least apparently)
4. Brief psychotic experiences, which are likely to have a paranoid

quality
5.Bizarre, dereistic, illogical or primitive responses on

unstructured psychological tests such as the
Rorschach, but not on more structured tests such as
the WAIS. (Characteristically, their interpersonal
relationships vacillate between transient superficial
relationships and intense, dependent relationships
that are marred by devaluation, manipulation, and
demandingness.)
The borderline patient represents a frequently encountered
type of patient, posing special problems for the psychotherapist and

the general physician and a thorny challenge to those interested in
the etiology and nosology of mental illness (Mack 1975). Painful lack
of agreement about the concept has been known for some time. A
serious attempt was made by Grinker and his co-workers (Grinker et.
al., 1968) using meticulous methodology, to try to better delineate the

borderline patient: It matters little whether we call the borderline
syndrome a disease, an arrest of development, an emotional

disturbance or a type of behavioral deviance. Likewise it is restrictive
to view the borderline from a single frame of reference such as the
biological, medical, psychological or social. The borderline, like health
and illness, is a system in process occurring in time: developing,

progressing and regressing as a focus of a large biopsychosocial field.
I think that we have to stay with this rather profound
conclusion also in the light of current knowledge on the subject.
Therefore, I have to warn the reader that if he is hoping from this
book to get the answers about the diagnosis, dynamics and treatment
of the borderline patient, he will be bitterly disappointed. The current
state of our knowledge on the subject does not permit a resolution of

the wide disagreement.
The best one can do at this time is to delineate the areas of

general agreement and indicate where there are unresolvable
differences. Summoning our patience, we then must look further at
the clinical material in the true tradition of psychoanalytically
oriented psychotherapy and hope eventually, by further and further

study and accumulation of clinical experience, to resolve the
disagreements. As Freud wrote (1914), I was not subject to influence
from any quarter; there was nothing to hustle me. I learnt to restrain
speculative tendencies and to follow the unforgotten advice of my
master, Charcot: to look at the same things again and again until they
themselves begin to speak.
The reader will discover fundamental and unresolvable
differences of opinion as to the foundations of psychic development,
and although we can discard a number of theories of basic

development as being too flamboyant or too far out or contrary to
common sense or reading like science fiction, there still is a
fundamental disagreement in the literature from at least two basic

points of view. This disagreement is not resolvable and leads to
entirely different theories of the development and treatment of the
borderline patient.
Although the term borderline appears from time to time in
psychiatric writing, major credit for delineating this concept and
making it clinically respectable goes to Stern. In three papers (Stern

1938, 1945, 1948), the first as early as 1938, he painted the clinical
and psychodynamic picture in broad outline and discussed special

problems in the treatment. Using the definitions of Freud in his
famous paper On Narcissism, Stern regarded narcissism as the basic
underlying character component of these patients, leading to the

development of an individual with certain typical personality
features. These in Sterns terms are: (1) psychic bleedingthe patient
goes down in a heap at each occurrence of stress in his life; (2)

inordinate hypersensitivitythe patient is constantly insulted and
injured by trifling remarks; (3) rigidity; (4) negative therapeutic
reactiona response of depression and anger to any interpretation,

which is experienced as an injury to a patients self esteem; (5)
feelings of inferiority and lack of self-assurance; (6) a tendency

toward masochism and wound lickingthat is, toward self-pity and
chronic depression; (7) a strange pseudo-equanimity, or outward

calm, which may be present, although not always, in spite of the
inward chaos; (8) a tendency to use projection, especially with people
in authority, and corresponding peculiarities in reality testing.
The study of the narcissistic neuroses, considered by Freud to
be among the most difficult to approach, has come into the

foreground in psychoanalytical research in the past fifteen years,
together with an increased interest in ego function and research in
the borderline psychoses and psychotic illnesses. Sterns (1938) first

paper is entitled Psychoanalytic Investigation of and Therapy in a
Borderline Group of Neuroses and is a very important study, giving a
clinical description as well as an attempt at a therapeutic approach to

what he calls the narcissistic character neuroses. According to Stern,
the narcissistic neuroses are closely related to borderline psychoses
and are in general unaffected by therapeutic methods successful with

the classical psychoneuroses. It is on the basis of narcissism, he
writes, that the entire clinical picture is built.
These patients suffer in the psychic field what Levy termed
affect hunger, and according to Stern this group never develops the
sense of security from being loved that is the birthright of every child.
Stern emphasized the borderline patients flatness of affect, in
contrast to Gundersons description mentioned above.
According to Stern, the patient cannot identify himself with the

analyst except through illusion. That is, the patient never identifies
himself with the analyst, but only with his conception of him through
a process of projection of his own ego ideal. It is this psychic figure

which talks to the patient. Therefore, if the patient is told, for
instance, that what he has just said indicates some suppressed
hostility from childhood to an older brother or father, the patient
literally collapses through fear of punishment by virtue of this having

been discovered.
In further describing the transference difficulties, Stern adds

that these patients very often react to a given interpretation with
chagrin, guilt, fear of punishment, and dread of not being approved.
All the patients energies are directed toward being approved and a
real awareness or ego integration of his own behavior does not take
place. Stern states, Actually, the investigation of the transference
phenomena informs us that as far as people go, these patients still live
in a world of their own childhoodso that getting well and being

adult are attained through wishing to be able to do what grown-ups
do, and this they dare not risk. In the imagination it is easy enough
and even while in the analysis, but independently the anxiety is too
great.
Stern summarizes by saying that a certain vagueness in the
paper is unavoidable because the material this group offered for
study ran so clearly in two directions: the psychotic and the

psychoneurotic. He insists that much more time and investigation are
necessary to evaluate the rather obscure phenomena these patients
present.
In 1945 Stern published an extension to his first paper entitled
Psychoanalytic Therapy in the Borderline Neuroses. This paper

dealt with a technical change in therapeutic approach, namely, having
the patient sit facing the analyst during the interviews rather than lie
on the couch. Stern emphasizes the importance of the patience and
love capacity of the analyst in affording the patient a corrective
experience of his childhood environment, which conspicuously lacked
libidinous giving. I remind the reader that this was thirty years ago.
Stern regarded the entire problem as a developmental injury
caused by the lack of spontaneous affection from the mother. Such

patients were describedas indeed they are described today, and
correctly soas traumatized preoedipal children suffering from a
profound affect-hunger. I thing Greensons (1958) screen patients
also fall in this category.
The second author to make an important contribution to the

subject was Deutsch (1965), who described the as-if personality,
which represents a subclass of the borderline patient group. In
general the as-if personality is an extreme caricature of Riesmans

(1955) other-directed personality. Although he appears outwardly
amiable, he has no identity of his own and is not capable of forming
any genuine emotional attachment to people or moral principles.
While there is a poverty of object relationships and again the central

issue seems to revolve around narcissism, no obvious defect in reality
testing is present; in fact, he may become a very successful politician
or administrator.
The subject of the borderline patient gained tremendous
current prominence with the introduction of new terms by well-

known and highly respected authors. The first of these was pseudo-
neurotic schizophrenia, introduced and investigated by Hoch and his

coworkers (Hoch 1949, 1959, 1962). Patients suffering from this
disorder are characterized by pananxietythey are made anxious
by everything conceivableand panneurosesthey present all
varieties of neurotic symptoms, shifting back and forth over our

nosological classifications. They may at times show clear-cut
psychotic manifestations and even psychotic episodes, but these do
not last and the patients as a rule do not deteriorate into chronic
schizophrenic psychoses. Hoch vigorously opposed including
pseudoneurotic schizophrenic patients among borderline patients; he
considered them a variety of paranoid or catatonic schizophrenics.
However, clinical experience and common usage have tended
to include pseudoneurotic patients among borderline patients,

because their pananxiety and panneuroses make it impossible to
classify them as either neurotic or psychotic and, more importantly,
because these conditions do not deteriorate into schizophrenia

indicating a certain remarkable stability to the condition. The
previously described narcissism and poverty of object relations of as-
if or borderline patients are typically present in pseudoneurotic

patients.
It is very important not to confuse the concept of borderline

patient with such cases as ambulatory schizophrenia or latent
schizophrenia, generally designating schizophrenic patients who are
not so sick as to require hospitalization. Thus, ambulatory or latent

schizophrenics show the typical symptoms of schizophrenia, except to
a less obvious degree; careful clinical examination may be necessary
to pick up the typical schizophrenic syndrome, and a diagnosis then
can be accurately established.
In many ways the evidence from clinical work militates against

including the borderline personality disorder on the genetic spectrum
that has been called the schizoid spectrum by a number of authors.
Thus, if there is a genetic or polygenetic diathesis to the schizophrenic

spectrum, it is different than any such diathesis involved in the
development of the borderline personality disorder, and indeed there
is currently no evidence that genetic factors are involved in the latter
at all.
Knight (Knight 1953, 1962a, 1962b) gave impetus to the
serious psychoanalytic investigation of borderline cases by discussing

them in terms of variable impairment of ego function. This provided a
partial theoretical explanation for the confusion in the nosology,
although his use of the term borderline schizophrenias again tended to

blur the distinctionwhich is a very important onebetween
borderline patients and ambulatory schizophrenia patients. Knight
explains that in the borderline patient the ego is laboring badly. The
superficial clinical picture of the variety of neurotic symptoms and so

on represents a holding operation in a forward position; the major
portion of the ego has regressed far behind this forward position in
varying degrees of disorder. The great danger to the clinician is to

misunderstand these forward holding positions as constituting the
illness and to attempt to treat them, whereas they actually represent
the healthiest part of the patients ego function.
Not only may the borderline patient show a variety of neurotic

symptoms, but he may show a variety of delinquent or acting-out or
pseudopsychopathic symptoms, involving him in all kinds of
difficulties with society. This would be logically expected if the
condition represented the impairment of ego function. Such patients,

for example, may involve themselves in all sorts of delinquent activity
at various times in their lives, but it is unusual to find them engaged
in any kind of brutal crimes.
In our era the most typical of these patients appear in the
general physicians office due to the syndrome of periodic
hyperingestion (Chessick 1969). To the despair of their physicians

and the panic of their families, these patients may consume large
quantities of substances or combinations of substances including

opiates, barbiturates, marihuana, meprobamate and other
tranquillizers, nicotine, mescaline, alcohol, amphetamines and other
psychic energizers and food. At other times there may be complete or

almost complete abstention.
Various physical and psychic symptoms may periodically
become intense; these include aches and pains, gnawing and weird
abdominal sensations, insomnia (sometimes very severe), anxiety
attacks, epileptiform seizures, tics and twitchings and, of course, the
symptoms of depression. Such symptoms are sometimes followed by
an explosion of hyperingestion in which the patient is functionally

partly or completely paralyzed and concentrates all his energy on a
compulsive stuffing in of various substances while other activities are

neglected. Substances hyperingested may vary from episode to
episode and the diagnosis of alcoholism or addiction may be
mistakenly made at this point. However, although the patient may

shift back and forth, he is on the whole able to function reasonably
effectively in society and he does not deteriorate.
The clinical delineation of the borderline patient (Chessick
1975) then should include the following characteristic features:
1.Any variety of neurotic or quasi-psychotic, psychosomatic and

sociopathic symptoms in any combination or degree
of severity may be part of the initial presenting
complaint. Either a bizarre combination of such
symptoms cuts across the standard nosology or the
relative preponderance of any symptom group is
constantly changing or shifting. Thus at least two and
preferably three diagnostic interviews at least a week
apart are mandatory in establishing the diagnosis, in
addition to a careful history taking including details of
all symptoms and their vicissitudes.
2. Vagueness of complaint or even a bland, amazingly smooth or

socially successful personality may be encountered.
Careful investigation in such cases reveals a well-
hidden poverty of genuine emotional relationships
behind an attractive and personable social facade.
Thus the patient may present either a very chaotic or
stormy series of relationships with a variety of people
or a bland and superficial but relatively stable set of
relationships. In both cases a lack of deep emotional
investment in any other person may be carefully
consciously or unconsciously concealed.
3. The capacity for reality testing and the ability to function in

work and social situations is not seriously impaired,
although the degree of functioning may vary from
time to time. On the whole, these patients are able to
maintain themselves, raise families and otherwise fit
into society. They do not present as drifters, chronic
hospital or long-term prison cases, totally antisocial
personalities or chronic addicts. On the other hand,
they have often tried everything and may present a
variety of sexual deviations, but they are not
functionally paralyzed by these or by their neurotic
symptoms or anxieties for very long periods of time.
4. These patients do not deteriorate. The borderline patient

suffers from a relatively stable and enduring
condition. He may endure what appear to be transient
psychotic episodes either for no apparent reason or as
a result of stress, alcohol, drugs, improper
psychotherapy and so on, but he does not remain
psychotic for long. He snaps out of it and often he
learns what will snap him out of it and administers a
selfremedy. At times this remedy may simply consist
of dropping out of an improper psychotherapy; at
other times it involves all varieties of ritual or bizarre
behavior. Sometimes his marital partner or friends
know about this and apply self-remedies for him
they consider this just his hang-up.
When the borderline patient is in one of his panneurotic,
pananxious, hyperingestive or psychopathic states, he causes

tremendous alarm in those around him and appears to be in a terrible
condition. At the same time he may frustrate all efforts to help at
that point, or if helped he may show a surprising lack of gratitude.

Borderline patients who suffer from various transient episodes soon
acquire a reputation in the family and are often rejected by physicians
as crocks or bad patients. They stimulate many unconscious and not
so unconscious maneuvers by both family and physicians to get rid of
them, for example, by sending them to a sanitarium for a rest.
A number of sociological authors have emphasized the major
change in presenting symptomatology found in the office of the

psychotherapist over the recent years. Nowadays, the presenting
statements deal with vague complaints of maladjustments and
discontentin short, they sound more like the borderline patient and
less like the classical neuroses described by Freud. The lack of
identity in these patients is linked by sociologists to the collapse of
institutional absolutes and values, allegedly leading to a sense of
futility, emptiness and longing. This problem is so serious in our time
that it deserves careful delineation, the purpose of the next chapter.
CHAPTER 2
Etiology
What are the factors in our rapidly changing Western society

which may spawn or facilitate the development of the borderline?
First of all, it is important to keep in mind that the complaints of the

borderline patient often resemble a caricature or exaggeration of the
complaints and behavior of the so-called normal people in our current
society. In fact, many as-if and other borderline patients are quite
successful in the superficial social and business world. This is in
marked contrast to the latent or ambulatory schizophrenic, whose

complaints are consistently more bizarre and who is usually an
unsuccessful person by societys standards of success.
Why is this a more difficult time to raise children than other

times? One way to look at this question is to look at other times.
Mirabeau wrote, The civilization of a people is to be found in the
softening of manners, in growing urbanity, in politer relations and in

the spreading of knowledge in such ways that decency and seemliness
are practiced until they transcend specific and detailed laws. . . .
Civilization does nothing for a society unless it is able to give form

and substance to virtue. The concept of humanity is conceived in the
bosom of societies out of these ingredients. What we are currently
immersed in is the opposite of civilization, and, in spite of how hard

individual parents may try, their children are caught up in the current
whirlwind of barbarism. This provides a counterforce from which
only the very strongest adolescents can emerge unscathed. Here are
some briefly described features of modern western culture that
interfere with maternal functioning and child development:
1. The breakup everywhere of family and family authority,

traditions and values, so that adolescents now more
than ever are more influenced by their peersand
usually the most malevolent of their peersthan by
their family.
2. The presence of a stable family, of a permanent and peaceful

marriage, is no longer even accepted universally as
desirable. In actual experience the adolescent from
such a household discovers he is in a dwindling
minority; if there is no fighting, no divorce, no
smoking and drinking in the house, the adolescent
almost feels a bit deprived and must seek or create his
own excitement.
3. In this era of future shock (Toffler, 1970) there is NO sense of

the future! Perhaps it can be argued that the aimless
hippie drifting of the flower child is a form of
preparation for a dark uncertain age with no future,
with atomic warfare impending, subject to
unpredictable shortages, economics disasters and
what else?
4. The destructive effect of the loss of grandparents and ancillary

family members such as uncles and aunts and so on
living in close proximity and playing a significant and
in older times often therapeutic role. The same can be
said for the loss of religious values and religious
figures and traditions and the constant moving of
homes with subsequent breakups of old friendships
and the necessity to keep new ones superficial in the
anticipation of moving again.
In discussing schizophrenia, Arieti (1974) points

out that in between the first edition of his book
(1955), and the second edition (1974), we have
witnessed in the United States a marked increase in
cases of schizophrenia occurring in adolescence and
early adulthood from the age of thirteen to the age of
twenty-three. He writes, We must assume that the
present cultural climate in the United States facilitates
the occurrence at an early age of that conceptual
attack to the self that brings about resonance and
unification with primary process experience. He also
stresses Riesmans (1955) discussion of the other-
directed culture. The models for youth are their
peers, their contemporaries, not the older generation
or the heroes of the past. As one borderline patient
told me, George Washington, Thomas Jefferson, those
memorials in Washington are now obsolete. The
conceived ideals are considered less distant and
expected to be more quickly attained, and
consequently the despair of the self occurs earlier in
life.
5. Affluenceour children usually have no personal experience of

deprivation or really hard times, which makes it
impossible for them to really imagine what hard times
are like and does not motivate them to provide against
deprivation.
6. Parents today live without values and without restrictions on

their own abuses of drugs, tobacco and alcohol. All
these are instantly and easily available; how can we
expect the child to deny himself when adults who
should know better do not and even spend a fortune
in advertising to urge people to take all kinds of
chemicals for every possible discomfort?
7. Insecurity and unsureness of parents as to how to raise

children and what to do. This is a relatively new
phenomenon in history, fostered by amateur
psychology, which abounds everywhere with
promises of instant solutions and instant relief for
everythingand every advice contradicts every
advice. Most destructive is the inconsistency of
mothering based on this confusion.
8. An age of high mobility and fast transportation; jet planes,

motorcycles and so on, with an emphasis on speed.
Mobility for the sake of mobility becomes a symbol of
independence and masculinity; only Ferdinand, the
queer bull, sits under the trees and smells the flowers.
Even girls run away and hitchhike if they are with
it.
9. Television: fast on-the-spot communication of sensational and

exciting news, the bloodier the better, and to fill in
the time between newscasts, the shows which make
up all sorts of violence and sexual sensation to keep
the viewer watching. Dedicated to only one purpose
to get the viewer to watch the commercialthere is
no level to which television will not descend and no
end in sight to the decadence. Thus our children are
bombarded from the earliest time of life with
primitive sensations, and this is their concept of
reality. What child today has the experience of
Bertrand Russell (Russell 1967) as he describes it in
his Autobiographylong quiet afternoons in his
grandparents large library, with nothing to do but
read and think?
10. The general decline of social values and the quality of life in
general, with corruption in fashion from the White
House down. To quote Nef (1967): Evil can be as
contagious as virtue. To the fast multiplication of
people and things, the mechanization and automation,
the hurry and crowding which art has had
increasingly to absorb or contend with during the last
hundred and fifty years, has been added the peculiar
fascination of doing wrong as an end in itself, at a time
when the word wrong has lost its meaning and
gratuitous acts of violence have acquired prestige.
It is clear from this brief description that the potential effect of

society in forming the borderline patient is both direct, in lethal
impact on the emerging personality of the child and adolescent, and
indirect, by serious interference with the mothers capacity to

respond to her infant due to a lack of her own sense of calm and
security in the maternal environment. The mothers of borderline
patients have been described as intelligent and overfeeding and as

mothers who were able to hide their anxieties and the emotional
impoverishment of their personalities behind pseudogiving. This is

combined with a stern and almost cruel often unverbalized demand
that the child live up to their expectations. This combination of
overfeeding and pseudogiving accompanied by the hidden stream of
demands produces in the childs mind a chaos that Leuba (1949) has
labelled deception, leading ultimately to severe defects in ego
development.
Based on the work of Marie Bonaparte, Leuba, the French
psychoanalyst, introduced a new concept in the study of narcissism,
namely the phobia of penetration, which means the fear of each living
cell and each living being of being penetrated and destroyed. This is
rather romantic but is interesting and useful clinically. He visualized
in this broader concept, which he called, biological narcissism, an
elementary fear and threat to the integrity of the body, the fear of the
cut-up body, and believed that all manifestations of this fear are
identical with the concept that Odier (1956) called primary phobias.
Just as one recognizes in all men the biological bisexual polarity, these
primary phobias are common to all beings and represent biological

narcissism.
The primary phobias embrace all phobic reactions of early
infancyfor example, the anxiety responses of a nursling to an

unexpected noise, a bright shining light unfamiliar to the babys eyes
or even a simple change in the dress of the nurse. Summarizing
briefly, these are elementary fears due to changes in the accustomed
environment and are experienced in the dim light of the ego function
of an infant as a terrifying unknown danger. Upon this elementary
core Leuba (1949) believed is added, without the infants being able
to establish any limit whatsoever, the concept of the fear of
penetration.
Turning to clinical material, Leuba quoted many cases of an

early dream of resistance to psychotherapy. Because of deep-seated
primary phobias and their weak ego structure, these cases either
pretend to get well or break up treatment after a few hours.
Secondary narcissism, according to Leuba, is a regressive

phenomenon and it is a defense against the recurrence of deception,
the fear of the feeling of abandonment arising from early, mostly
preoedipal childhood experiences. The point of departure for
narcissistic regression is a feeling of frustration. In order to soften
and despoil the armor of narcissism, treatment for Leuba, consists
first in affording a corrective experience of being loved in spite of the

deeply repressed aggressions engendered by the fear of
abandonment. This permits the person of the therapist to be afforded
a reinvestment in original libidinal drives that were so frustrated by

the deceiving objects of infancy.
Odier (1956) made an early contribution to the understanding
of these patients with his concept of the neuroses of abandonment.
He sees the anxiety of these patients as directly proportional to the

amount of insecurity in early childhood, producing regression to the
prelogical stage of infantile thinking. He describes this magic thinking
in detail as involving either (a) objectification of fearWhatever

threatens me is wicked, whatever protects me is good; (b)
objectification of anger toward animistic malevolent objects as
chosen; and (c) identification with the aggressor. The objectification
is the magical defense placing the anxiety and fear and anger
outside of the psyche onto external objects, as in phobias, or onto
fantasy objects, as in nightmares or religion. In a much later
publication (Chessick 1972b) I have described in detail the process of

externalization and existential anguish presented by the borderline
patient. The description is based on the loose concepts of Odier.
In the neuroses of abandonment, the anxiety is objectified onto
a human being instead of a cosmic image or a transitional objecta
human being who is then given the power of creating or abolishing
abandonment, insecurity and helplessness. This individual is seen as
all-powerful, sometimes benevolent and sometimes malevolent. In
this situation, the oscillation between love and hate, security and
insecurity, dependency and paranoia, and the rapid transitions from
euphoria to depression, all as a function of minor provocations or
reassurances from the chosen object, lead to the typical picture of the
borderline patient and the all-too-common phenomena described as

narcissistic object relations.
These concepts of deception and fear of penetration are
absolutely vital to an understanding of and empathic feeling for the

borderline patient!
Modells (1968) book Object Love and Reality is one of the

important monographs on the borderline patient and must be read
carefully. He bases his thinking on Winnicotts concept of good
enough mothering, which I have also used repeatedly in my books on

psychotherapy (Chessick 1969, 1971c, 1974b). He points out that
autonomous structures will be impaired if there is an absence of
empathy in the maternal environment, and the central theme of his
whole monograph is that the acceptance of painful reality rests upon

the same ego structures that permit the acceptance of the
separateness of objects. Thus the ego structure whose development
permits the acceptance of painful reality is identical to the psychic

structure whose development enables one to love maturely. He
writes: In both instances the signposts that indicate whether or not

such a successful historical development has been traversed is the
sense of identity. If one is fortunate enough to have received good
enough mothering in the first and second years of life, the core of a
positive sense of identity will have been formed. This core permits the
partial relinquishment of instinctual demands upon the object and in
turn permits the partial acceptance of the separateness of objects. It is
this process upon which reality testing hinges.
I strongly recommend that you review carefully Modells
monograph, because it is a very fundamental starting point in

reaching a deeper knowledge of the borderline patient. Earlier,
Modell (1963) stressed the importance of a core of positive sense of
identity, a sense of beloved self, which develops in infancy as a

response to adequate mothering. Without this sense of inner
sustainment (Saul 1970), thinking remains magical and object
relations remain primitive.
Ferenczi (1950) wrote a major paper in 1913 on the stages in
the development of the sense of reality. He delineated a series of

phases that the individual goes through, named a period of
unconditional omnipotence, a period of magical hallucinatory
omnipotence, a period of omnipotence by the help of magic gestures,
an animistic period, and a period of magic thoughts and magic words.

In all of these narcissism is characteristic, as is the domination of the
pleasure principle. There must take place the replacement, Ferenczi
points out, to which we are compelled by experience, of the childhood

megalomania by the recognition of the power of natural forces. This is
the essence of the development of the nature ego and leads to what
Ferenczi calls the stage of objectification, the ascendency of the reality
principle. Thus the first five stages persist only in fairy tales and
mythology under normal conditions.
This is developed further in an important paper by Murray
(1964). He describes the deep narcissistic sense of entitlement that
pervades the thinking of borderline patients which of course indicates

that the patient has not reached the stage of objectification. He
explains that the most important battle in the cure of these deep-
seated character neuroses, as he calls them, lies in transforming the

narcissism, that is, transforming neurotic attitudes based upon
persistent early narcissism into the adequate, appropriate and mature
ego-ideal orientations of adult life. He explains that the patient lives in

the narcissistic world of omnipotence with its unlimited power of
magical thinking and unlimited entitlement to the lust and
destructions of pregenital excitements. If this is not given up, then
the patients life will be critically limited, circumscribed and

indefinite. Therapeutic achievements are only passing in this
situation. This paper, in my opinion, presages what we will discuss in

great detail later, described by Kohut (1966) as transformations of
narcissism, and therefore is of historical importance as well as clinical
usefulness.
It is necessary to quote one more author in bringing the
historical review of psychodynamic descriptions of the development
of borderline patients to the place where we can consider

contemporary work. Gitelson (1973) wrote in 1958 on ego distortion,
which he considered fundamental in these kinds of disorders. He

suggested that the cases of ego distortion represent a way of life, a
method of adaptation demanded by peculiar or particular deviations
in the internal and external environment. The interplay of the various
factors cause what he calls inadequacy, hypertrophy and/or atrophy
of ego functions as such and produces an apparent imbalance in the
picture of the total ego. He explained that such patients have
encountered unusual stress in their relation to their original objects
centrally the motherand the consequences are seen in pregenital
disturbances in the economy of the libido and aggression, in defective

superego development and in compensatory internal and external
adaptative and adjustive accommodations of ego function.
Gitelson feels that these syndromes should be thought of as

narcissistic personality disorders. Already at this point it should be
clear that anyone who is going to understand the borderline patient
will have to have a thorough understanding of the vicissitudes of
narcissism and of psychoanalytic ego psychology, because these

patients represent in many ways the most subtle and complex
malfunctions and maldevelopments of psychic structure of any

patients that we deal with.
Kernberg (1967, 1968), for example, stresses the borderline
patients lack of anxiety tolerance, lack of impulse control and lack of

sublimatory channels, and he contends that oral aggression poses a
crucial role in the psychodynamics. There is a premature
development of Oedipal conflicts as an attempt to escape from oral

rage, with a subsequent condensation of pregenital and genital
conflicts. In other words, there is fundamentally a pathology of

internalized object relationships and an intensification and
pathological fixation of splitting processes in the ego functions of
these patients.
Even the descriptive difficulties in any discussion of the
borderline patient get more and more confusing as one reads more
and more of the literature. For example, take the so-called good or
bad hysteric (Lazare 1971). The true good hysterical character is
the person who is supposed to have quick and intuitive resonance
with others, be a sexually exhibitionistic, competitive, buoyant and

energetic, ambitious, histrionic individual with a strong sense of guilt
who does well in his work. Such a person is not promiscuous,
although he is not able to make a major sexual investment in others

and he often is involved with an unavailable partner. He shows a
predominance of Oedipal conflicts, retains significant ties with old
parents and is relatively well integrated in ego and superego
functions. This is contrasted in the literature to the borderline

hysteric or the bad hysteric or the hysterical personality, which is
manifested by low self-esteem, a sense of helplessness, whining and

contrariness, inconsistent and irresponsible work habits, rapid shifts
between intensive negative and positive feelings, sexual promiscuity,
drifting relationships, a poor tolerance of frustration stress and the

predominance of infantile oral conflicts. His mental life is
characterized by polymorphous perverse fantasies and few conflict-
free areas, and there is emphasis on the defense of splittingthat is
to say, there is an incapacity for synthesizing good and bad introjects

and identifications. We will have a lot more to say about the latter
problem further on.
This represents a very confusing situation in the literature, and
there is a lot of argument, as, for example, in the panel report by
Rangell (1955), where much of the panel ended up in a discussion of
whether borderline states exist at all. Some very highly respected
people, such as Gregory Zilboorg, insisted that there is no such thing.
At any rate, it was admitted in this panel that we are not sufficiently
expert to judge by any one measure the psychological status of a
borderline case, except at least to some extent intuitively and

impressionistically.
CHAPTER 3
State of the Ego
The group reported by Rangell (1955) felt that such concepts

as ego weakness tell us little. For example, an ego may be strong in
one capacity and weak in another. It may have a great capacity for
integration while remaining weak in regard to certain cultural
realities, and so on. This group feels, to quote Ranged, The essential
pathology is in the ego functions, with regression and primitivization
of these. The primary process invades the secondary, and there is
difficulty in separating a judging ego from the experiencing ego. The

implication of this for treatment is that it becomes necessary to tell
such a patient not only, you are reacting as if I were your father, but
also to add, and I am not your father, thus helping him acquire a
judging ego. I think this is an important clinical pearl.
The issue of the ego state in the borderline condition has been
reviewed by numerous authors, and it sometimes becomes rather

mystical. For example, Cary (1972) insists that he can differentiate
the borderline condition by what he calls a structural-dynamic
viewpoint. He claims that the borderline structurally represents an

arrest of early development when ego and object tended to be fused.
From the structural-dynamic viewpoint, according to Cary, there is
basically a fear of the loss of a unified ego sense and a fear of a loss
of relations with others, leading to an alternating struggle between
the wish for fusion of ego with object and the opposing wish for
separation of ego from object. This is a rather complicated way of
describing a very common phenomenon that we encounter in the
borderline patient having to do with the sense of self and the
traditional need-fear dilemma.
What do we really mean when we speak of ego weakness or
ego defect in the borderline personality disorder? The answer to

these questions as they emerge from clinical experience in the
treatment of borderline patients throws considerable light on the
increasingly complex notion of the ego and its functioning. Knowledge

gathered about the operations of the ego for the treatment of such
patients is applicable to the treatment and understanding of all
patients, just as in the old days information about derivatives of id
processes as they appeared in raw form in the productions of

schizophrenic patients was useful in investigating the unconscious of
all types of patients regardless of the intactness of their defensive

structures.
The whole concept of the psychoanalytic assessment of ego
weakness is a very difficult one. For example, according to Zucker

(1963), in such assessment a number of personality areas tend to get
overlooked but are very important in detecting ego weakness:
1. The disturbance in screening activities with a reduction of

sensitivity to external stimulation as
overcompensation for inadequate boundaries
between outer and inner reality
This is an interesting clinical point. For example,
people with ego weakness at times simply do not
hear what you are saying. They ask you to repeat
yourself or they get confused in directions if they are
going somewhere on a train or the bus. They dont
notice the external stimuli they should be noticing and
there seems to be a higher threshold at times for
stimuli to register.
2. The fusion of different realms of cognitive experience
3. The tendency to multiple identifications
4. The phenomenon of the fluctuating body image [This in my

experience is especially noticeable in borderline
adolescents.]
5. The problem in segregating the consequential from the

inconsequential [This is a good one to look for
clinically.]
6. The extension of the ego field into other fields or entities
Some of these items may not be entirely clear so I will go into
them a little more. In talking about the fusion of different realms of
cognitive experiences, I think Zucker is talking essentially about

microscopic difficulties in keeping associations orderly and not
contaminating each other. He gives as an example the following
statement of a patient: The subway seemed to develop a terrific
speed todaythat is why I myself like to do things in a hurry. He

sees this kind of logic as based on the persons inability to remain
within a single frame of reference.
According to Zucker, extension of the ego field to other fields
or entities is a peculiarity which shows up most distinctly in testing
behavior but also can be observed clinically. He explains that the
description pertains to the type of subject who in a varying degree

encounters problems in keeping apart his inner reality from outer
realities. For example, a patient may say he is afraid to die because he
is convinced that with his death the world may come to an end too, or
he may have difficulty in distinguishing between his daydreams and
reality.
This is all on a microscopic level and it is similar to Knights
(1962) description of the borderline schizophrenic mentioned earlier.

How do we distinguish ego weakness in the borderline patient from
the borderline schizophrenic or ambulatory schizophrenic? Let us
begin with DeWald (1964) on the subject of ego strength and ego
weakness. Generally an individual who is capable of maintaining
various activities, work and other relationships, in spite of his
disturbances, and capable of meeting and dealing with the

vicissitudes of his life can be said to have some ego strength and will
successfully deal with the stresses of insight therapy. DeWald points

out, The more an individual has had a general pattern of persistent
effort in a goal-directed fashion, and of success in the various
ventures that he has undertaken, the more likely will he be to sustain

his effort during the course of the treatment, and ultimately to
achieve some measure of success.
At the other end of the spectrum, of course, the person whose
life pattern has been one of repeated failures, ineffectual adaptations

and major disturbances and disruptions will repeat these ego
patterns in the therapeutic situation, and the patient may even be
incapable of using what insight he gets to effectively modify his
previous patterns of disturbance by virtue of his inadequate overall
ego capacity.
In addition to this, from clinical study it is apparent that
certain specific ego functions have very important impact. Obviously,
the patient who is able to make significant investments in other

people will do better in psychotherapy. The capacity of the patient to
make a sustained object relation over years of time is very important.
By virtue of rigid defenses, some individuals are simply unaware of

their inner emotional impulses, conflicts and reactions and insist on
focusing everything on external, current, realistic events. Such
patients will have great difficulty in uncovering psychotherapy; at the
other extreme, patients who are morbidly introspectiveto the point

of excessive continuous rumination with so much internal awareness
that they exclude reality in real life circumstancesare people with

serious ego weaknesses who will have problems in psychotherapy.
There are certain specific defense mechanisms which when
used extensively suggest ego weakness and poor prognosis in

uncovering psychotherapy. DeWald mentions reliance on projection,
massive denial, major withdrawal and so on. Especially difficult is the
acting-out patient who deals with his psychic conflicts primarily that
way; the prognosis is even worse if acting out is combined with
projection. At least average intelligence is required in uncovering

psychotherapy, although I would not say that a poor intelligence is
quite the same kind of ego function as the others under discussion for
purposes of evaluating whether a person is borderline or not. The
capacity to tolerate anxiety and frustration while working toward a
long-term goal is an important measure of ego strength or weakness
as is the willingness to postpone immediate gratification in the hope
of achievement of long-range goals.
To conceive of ego weakness simply as consisting of a rather
frail ego barrier which, when assaulted by id derivatives, is unable to

prevent them from breaking through and flooding the ego appears in
the light of modern clinical experience to be superficial and
insufficient. Today we conceive of the ego as an overall structure in

which substructures determine specific functions as well as being
determined by each other. Thus ego weakness should be
conceptualized not simply as absence or weaknesses in such
structures but as replacement of higher-level by lower-level ego

structures. For example, Kernberg (1975a) explains, a typical feature
of ego weakness in borderline patients is ... evidenced by the lower

defensive organization of the ego in which the mechanism of splitting
and other related defenses are used, in contrast to the defensive
organization of the ego around the higher mechanisms of regression

and other related defenses in the neuroses. Typically the ego
function in borderline patients is characterized by-splitting in which
internalized object relationships are split into good and bad and there
is an emphasis on introjections and projections with externalization.
This view of Kernbergs is considerably more tricky than

appears on the surface. Kernberg (1973) emphasizes levels of
internalized object relationships. These are: (1) a basic primitive
level characterized by multiple selfand multiple object
representations, corresponding to primitive fantasy formations
linked with primitive impulse derivatives; and (2) a higher level of
internalized object-relationships, characterized by sophisticated,
integrated self-images and by sophisticated, integrated object-images
linked with higher levels of affect-dispositions. All of these higher-
level object relationships reflect the early childhood experiences and

conflicts between the individual and his parental figures and siblings.
In addition to this, the differentiation of the transference in the
borderline patient from that of the transference in the neurotic

patient is that the transference neurosis in the neurotic patient is a
more realistic, dyadic, Oedipal-triangular, or sibling relationship,
while in the borderline patient, primitive object-relationships are
activated with multiple self-images and multiple object-images

representing the deepest layers of the mind. Now most authors are
agreed on the phenomena described here, but they are not in

agreement on the explanation of these phenomena. The primitive
multiple object relationships are of a fantastic nature and do not
directly reflect actual past interaction with the parents as do the

higher-level transference reactions in the neurotic. Rather, according
to Kernberg, they reflect early fantasy structures, that is, fantastic
relationships to inner objects which are normally submerged within
the structure of more realistic transference dispositions in the context

of an integrated strong ego and superego. Actually, even underneath
what appears to be fragility in these patients there are often
extremely rigid, primitive and pathological ego structures; anyone

who has attempted psychotherapy with such patients can attest to
this!
Sadow (1969) proposed a schema that utilizes the central role

of the ego as the axis of a continuum along which are located the
psychoses, borderline states, transference neuroses and conflict-free
capacities. He presents a classification of the emotional disorders

based predominantly on a view of the patients ego functioning,
thinking of the ego as a relatively fixed group of functions within the
personality, composed of both conscious and unconscious parts, and
comprising a variety of attributes and skills which vary in

composition and degree from individual to individual.
This ego axis represents the ego in a sequential line of
development from earliest infancy to full maturity. Movement along
this axis is depicted as a regressive shift or a progressive shift. Thus

the successful interpretation of and working through of a neurotic
conflict would result in a movement toward the conflict-free zone of
ego functioning. Ego depletion, whether based on illness, fatigue,

neurotic conflict, object loss or so on, would be depicted by a
movement in a regressive direction. Obviously, the entire range of
psychopathology can be superimposed on the axis, with very
regressed psychotics at one end and healthy sublimated personalities

with lots of conflict-free ego function at the other end.
In investigating the position of borderline patients on this ego

axis, one of the most important clinical facts that emerges is the
capacity of these patients to have a tremendous range and flexibility
of movement along the ego axis. It is clear that in any evaluation of a
patients ego we must study the ego operations not only to discern
higher or lower levels of predominating defensive processes, but also
to discern the egos capacity for motility back and forth on the ego
axismotility in which an oscillation occurs from nonstructuralized

regressive organizations at one end to conflict-free functioning on the
other. A great many therapists have been fooled into a pessimistic or
hopeless decision about the prognosis for the treatment of a

borderline patient because they observed the patient during a period
when he was temporarily residing in the regressed area of the ego
axis. Thus many borderline patients can sometimes function quite

well and at the same time in other situations or during certain other
periods appear to be psychotic and even hopeless.
The borderline patient can also fool us because under the
regressive pull of drives and defenses, certain autonomous ego

functions may become what Lowenstein (1972) calls
reinstinctualized so as to act in the service of resistance. The specific
medium through which we observe the mental apparatus is the
autonomous ego of the patient. The interactions and conflicts in

which we are interested take place not only between ego, id and
superego, but also within the ego itself. Hartmann (1950) made the
famous extension of the concept of conflicts from the well-known

conflicts between drives and defenses to conflicts within the ego
itself. Hartmann distinguished between the intrasystemic conflicts
within the ego itself and the more traditional intersystemic conflicts.
Autonomous ego functions which become reinstinctualized
appear on the surface to remain autonomous unless the therapist is
extremely careful in his evaluation. For example, the use of fancy
psychoanalytic terminology by a professional patient in therapy such
as a social worker, psychologist or psychiatrist can fool the therapist
into thinking that an autonomous function is being utilized in which
the patient is observing and understanding himself in a very scientific
mannerwhereas actually there has been a reinstinctualization of
speech in the service of resistance.
On the same subject of intrasystemic conflicts, it is important
to note that typically in borderline patients mutually incompatible
character traits may alternate as an indication of the extent to which

conflicting identifications have been integrated into the character
structure and tolerated by the ego or superego. Of course, a close
correspondence exists between the levels of structural organization
of the ego and of the superego, since related vicissitudes of internal

object relationships determine both ego and superego pathology, and
of course, borderline patients show frequent irregularities in

superego development. Though frequently superego formation in
borderline patients is superficially surprisingly good, the
characteristic of alternating between subservience to conflicting ego

ideals is a typical sign that underneath an apparently smooth
superego functioning is a lack of consistency and integration within
the personality.
Thus in the clinical evaluation of the patientany patientthe
concept of the ego state of the patient must be very carefully

examined and assessed (1) from the point of view of the
predominance of higher-or lower-level (primitive or less primitive)
sets of ego operations; (2) in terms of the capacity for mobility along
an ego axis in which, for example, the ego of the borderline patient
shows a remarkable capacity to shift back and forth from very
regressed states to autonomous ego function; (3) from the point of
view of the patients capacity for autonomous ego functioning, which
is a necessity for any uncovering psychotherapy (this must be
carefully distinguished from reinstinctualized situations in which the

ego functioning is apparently autonomous but actually in the service
of resistance); and (4) for the relative presence or absence of
intrasystemic conflicts leading to what Erikson has called identity

diffusion.
I am often asked about the value of psychological testing in
borderline patients. There have been a number of comments on this
in the literature, but there is no clear-cut pattern of psychological test

results established as related to the diagnosis or detection of the
borderline condition. According to Pfeiffer (1974), the value of

psychological testing lies in delineating the area of current concern,
conflict or symptomatology rather than in establishing the specific
diagnosis of a borderline state. He suggests, for instance, that the

MMPI and various kinds of depression or anxiety scales are distinctly
useful in establishing the existence and severity of target
symptomatology, as he calls it, and similarly, he suggests the
Rorschach and TAT to identify areas of concern or conflict.
More traditionally, the Rorschach test has been used to

provide a measure of the patients adequacy of reality testing in a
descriptive sense. The traditional description in the literature is that
borderline patients on unstructured tests such as the Rorschach show
worse performance than the neurotics do, but on structured tests, like
the MMPI, they do as well as neurotics. No characteristic or reliable
descriptions of borderline psychological test results exist in the
literature at the present time.
Therefore, the diagnosis of the borderline condition or
borderline patient has to remain essentially a clinical diagnosis on the

basis of careful clinical evaluation of the patient. Pfeiffer (1974)
quotes Vaillant, who attempted to make a kind of classification for the
defenses used as they are observed clinically. Level I is described as

the narcissistic (which is not the proper term); what Vaillant called
the narcissistic defenses include delusional projection, psychotic
denial, distortion or depersonalization and so on. Level II defenses
include projection, denial through fantasy, hypochondriasis, passive

aggressive behavior and acting out. Level III defenses, called
neurotic, involve intellectualization, rationalization, repression,

displacement, reaction formation and counterphobic and dissociative
reactions. Finally, the so-called mature or Level IV defenses are
altruism, sublimation, humor, suppression, avoidance, anticipation

and conscious control. This classification, it should be noted, mixes
several levels of discourse and is metapsychologically unsound.
At any rate, it is the fluid use of the variety of defense

mechanisms which characterizes the borderline patient. Thus, the
patient clinically shifts back and forth with great mobility among all
these levels of defenses, so the therapist has to evaluate clinically not
only the specific defenses the patient uses and how they are different
at different times, but also the ease with which the patient shifts back
and forth between these defenses and defense levels. This is
essentially a clinical skill and must be based on a careful set of
diagnostic interviews.
CHAPTER 4
Typical Borderline Complaints
For the past twenty years I have been frequently confronted in

my office practice in psychotherapy with a rather typical patient as
indicated by the following case vignettes. The first patient is a thirty-

five-year-old woman who has reached a certain degree of executive
success in a well-known corporation. She is respected by those
around her. Her efficiency and industry are often without peer and
she is responsible for the production of a number of items, with large
sums of money hinging on her executive decisions. Yet she seeks

therapy because, although she functions so well, she has vague
complaints of restlessness, dissatisfaction and of being alive but not
alive. Her marriage has failed. Her husband has turned from her to
some perversion, which surprised her entirely. Her later relationships
with men have been characterized by mere sexual promiscuity, by

utter lack of emotional attachment. Again and again she has started
out to form a relationship with this man or that and again and again it
has deteriorated into mere technical sexual prowess.
The patients life consists essentially of two phases of existence

which alternate with each other. Either she is crying herself into a
state of exhaustion and then enjoys the relatively quiescent feeling
after such catharsis or she is getting somebody to hold her, for which
she is trading sexual relations. The holding provides a magical
sensation that everything will be all right. She remembers as a child
being in bed with her sister and insisting I held on by putting an arm
around her sister in spite of her sisters repeated protests. Only while
she was holding on in this manner could she go to sleep. There was
nothing sexually arousing about this and no sexual play of any sort
took place in this situation.
Here is a similar vignette: A twenty-seven-year-old nun has

spent the last six years in a convent. She reports that she went about
all her duties in exemplary fashion and with continual praise from her
superiors. She was thought of as a pious dedicated, hard-working,

reliable and mature woman. She followed all the rules correctly.
There was never any serious problem.
At the same time the patient felt continuously that she was
dead. She went through the motions of living always with a strange,
almost indescribable feeling that she was not alive. This went on and
on, and it seemed to the patient that it would always go on. She never
consciously worried about it. She felt that somehow this was her
perpetual fate, the background of her existence.
One of her superiors began to seduce her. She was attractive
and young and this superior, who had a history of homosexual acting
out, began to encourage the patient to come to her room and caress
her. The patient refused to do any active caressing but did allow
herself to be caressed, and a series of clandestine evening meetings

began to take place in which the patient and the superior would climb
into bed and the superior would hold the patient close and caress her
breasts. In spite of increased and impassionate urgings and pressures
from her partner, the patient never allowed this to progress to
anything further.
This stirred up a tremendous conflict in the patient because

she felt that what she was doing was really not in her best interest,
although the superior kept describing the procedure as therapeutic
for her, and indeed, the patient noted with surprise and astonishment
that while she was being caressed and held, the feeling of being dead
would go away! Finally the patient went to a higher superior and told
the whole story. To her surprise she was asked to leave the convent
and nothing was done about her seducer, who appears to have made a
large financial contribution to the order.
The sudden experience of being held and caressed caused an
explosion inside the patient. She did not become psychotic, but she
could no longer accept as her fate the perpetual feeling of being dead.
This is in interesting contrast to the schizophrenic patient, who might

at this point develop a homosexual panic. She began to realize she
was missing something that she could have and she began to strive
for it. This was her chief reason for coming into therapy: I feel dead
all the time. I know I dont have to feel dead and I want to do
something about it.
What were these women looking for? What was wrong? They
were not overtly psychotic, and they were not neurotic as far as
traditional diagnostic categories are concerned, but clearly they were

severely disturbed individuals with deeply abnormal interpersonal
relations. Their histories also contained serious transient
psychosomatic disorders.
Hollender (1970; Hollender et al. 1969, 1970) has written
three papers on the need to be held in women. He reports that for

some women in his series, the need or wish to be held was so
compelling that it resembled an addiction. Body contact commonly
provides feelings of being loved, protected and comforted: The need

or wish for it is affected by depression, anxiety, and anger. . . both
direct and indirect means are used to obtain the holding and the
cuddling desired. Sexual enticement and seduction are common
indirect means. But it is not really clear at all what being held does
for these patients or what is really wrong. In repeated cases such
patients (male and female) have begged significant others (and
sometimes also me) even simply to hold their hand, because they
insisted that without the tactile stimulation they just do not feel alive.
It is not a sexual desire and it does not seem to be directly related
primarily to infantile wishes for tactile pleasurealthough certainly
this is a component of it. There seems to be an additional component

in which some sort of profound sensation of deadness can only be
neutralized by the physical touching presence of another human
being and not by anything else, according to the patient, including

talk, psychotherapy or interpretations of any kind.
After a few cases of this sort, I began inquiring more carefully
into my patients habits regarding tactile contact and I found to my
surprise that quite a number of patients with predominantly
existential complaints (such as difficulty in finding meaning in life,
vague restlessness and discontent, a sense of boredom or what Sartre
would call nausea and so forth) often displayed in one or another way
an extraordinary need to be held or for tactile contacteven with a
transitional objectwhich could not be replaced by anything else.
(To anticipate myself, in later discussion I wish to insert the
serious warning that for the therapist to provide such tactile
stimulation under any guise whatsoever is very destructive for both

the patient and the patients therapy, and nothing that I am saying
should be interpreted as implying that this is ever useful in the
treatment of any patient in psychotherapy.)
Perhaps the most popular and well-known problem presented
by borderline patients in office psychotherapy today is that of so-

called existential anxiety. Here is an example: A. P. was a handsome
twenty-nine-year-old man who began therapy in a state of almost
psychotic collapse. Until 1962 he had been a brilliant and promising

graduate student at a major university working for his Ph.D. He was
the favorite of the department and even took over a professors
course while the professor was away on a tour.
As the time for finishing his Ph.D. approached, A. P. began to
feel overwhelmed with the feeling that life made no sense, that it was
absurd, that everyone and everything was a fake, and that all people
were soon to die anyway. Suddenly one day in the middle of teaching
a class he simply walked out of the classroom, packed his suitcase and
left school. Not long afterward he got married and worked as a
salesman.
During the marriage he had spells of silent withdrawal and
depression during which he seemed almost catatonic. He was unable
to concentrate or do anything during these spells. At other times he

would have fits of uncontrollable rage and would break the dishes
and kick out the paneling on the door. Needless to say, the marriage
did not last and a divorce was obtained. In desperation he sought out
psychotherapy, although he had few financial means. I saw him
essentially as a charity case through a community agency; complete

physical examination by several physicians revealed that he was
without organic disease.
Even in the diagnostic interviews he displayed the same

oscillation of moods described above. In addition, a flood of bizarre
dreams and fantasies plagued him, along with homosexual
preoccupations. His history showed many attempts to escape the
dreary boredom and nothingness that he felt by travelling from city to

city. He wrote to me, I dont think you have any idea of the
desperation I find myself in. Only to split (go somewhere else) is to

give in to the ever present hope that somewhere life may be better. I
dont know how much longer that might be the case, and when I
realize that life is no better anywhere than here, thenI am begging

you to show me something to make it worthwhile.
This sense of deadness, emptiness, nausea, anxiety, dread and

lack of meaning in a dreamlike life is perhaps the most dramatic and
common presenting symptomatology of the borderline patient, and it

is often presented in an artistic or very articulate manner with very
appropriate affect.
The immediate impact of this on the therapist is to call forth
the notion of relationship. Clearly the relationship between the
therapist and the patient has the most obvious and important
dramatic influence on these kinds of complaints, and everyone knows
this. But it usually does not have a lasting influence nor does it lead to
any kind of basic changes in the patient. In fact, the whole question of
how you bring about change in psychotherapy remains an
exceedingly debatable one today.
To be more specific, three major debates are going on in our
field on this subject, and these debates tend to be stirred up every
time a discussion of a treatment of a borderline personality disorder

takes place. In the first debate, one side argues that certain factors are
crucial and common to all forms of intensive long-term
psychotherapy, whereas the other argues that forms of intensive
long-term psychotherapy can be distinguished on the basis of which

factors are crucial to their success. In the second debate one side
argues that the emotional interaction, the real object relationship or

the unconscious interaction between patient and doctor is crucial in
all forms of intensive long-term psychotherapy regardless of the
rituals or theory or school that is employed. The other side maintains

(a) that there are some forms of intensive long-term psychotherapy,
for example supportive psychotherapy, where the emotional
interaction is crucial, but (b) there are other forms where it is minor
and insight is the crucial factor in success.
In the third debate one side maintains that there is no basic

distinction between psychoanalysis and psychoanalytically oriented
psychotherapy, as they are both forms of intensive long-term
psychotherapy with common crucial factors. The other side argues
that there is a distinction between the pure gold of psychoanalysis
and the copper of direct suggestion as follows: In psychoanalysis
insight through interpretation of the transference neuroses is
crucial, whereas in suggestionall other psychotherapythere is
merely education or inspiration, etc.
In the most modern form of this debate, one side maintains
that an existential encounter of some sort with the patient is vital to
success in the psychotherapy or psychoanalysis, whereas the other

side maintains that only if the patient can form a workable
transference neurosis and undergo a careful and meticulous analysis
of this transference neurosis can there be a major and lasting change
in the personality of the patient and his disorder.
These issues will repeatedly come up in this book, and the

reader who is hoping for a solution or an answer to them is going to
be disappointed. There simply is no agreement on the issues raised in
these debates, and I will adopt a middle position, which will be
implicit in the various descriptions of the disagreements about the

psychodynamics and psychotherapy of the borderline patient that
will appear here. As I indicated at the beginning of the book, there is
no set answer to some of these questions about the borderline patient

and the reader will have to pick and choose and test clinically as he
sees fit.
A great reward from the psychotherapy of the borderline
patient is that it forces us to reexamine carefully and constantly every
aspect of the psychotherapy process as well as every aspect of
ourselves. No other type of patient so persistently, lucidly, and
dramatically challenges our personalities and also our therapeutic

procedures and convictions. The borderline patient provides a
continual test of our capacity for empathy, of our frustration
tolerance, and of how well analyzed we are and how well our infantile
narcissism has been integrated into our adult personality structure.
These patients sharply spotlight confusions or inconsistencies in our
convictions about psychotherapeutic process and offer a unique

opportunity for intellectual, professional and personal expansion.
Part II
METAPSYCHOLOGY
CHAPTER 5
Basic Metapsychology
The metapsychological characterization of the borderline

patient is unclear and controversial. There is no doubt that Freuds
paper On Narcissism in 1914 represents the starting point. In this

paper he discusses how the ego invests both the objects and the self
with libido, and he mentions, Narcissism in the libidinal aspect of
egoism. (Egoism is defined here as self-regard or selfishness.) This is
carried further in his papers on Economic Problems of Masochism
(1924) and Mourning and Melancholia (1917). In the latter he

describes how the representation of the lost object is invested with
libido formerly invested in the object. Thus the individual loves the
memory traces instead of the lost object and in this case the ego has
been invested with libido.
Even on basic definitions and word usage there are profound
differences of opinion in the literature. As an example, see the review

by Pruyser (1974), who discusses the concept of splitting as it is used
in psychoanalysis and psychiatry. In forty-six pages he reviews this
termwhich is too slippery and too hard and probably should not be
used at allas it is used in a different way by almost every famous
writer in psychiatry, writers such as Bleuler, Freud, Fairbairn,
Hartmann, Guntrip, Jacobson, Jung and others. Terminological

confusion is rampant in our field, so I feel it necessary to call
attention to certain basic definitions.
There is enormous debate about the concept of the ego. The
English school of object relations theory, especially Fairbairn and
Guntrip, reject Freuds basic ideas of the ego. Instead Fairbairn (1952)
postulates that at birth there is a pristine, unitary, whole human
psyche with ego potential that immediately begins to grow into a
developing self, a person-ego. There is no impersonal id; all is ego,

and if the infant experiences absolutely good object relations,
development could proceed as a stable, unified, steadily early-
enriched growth of the pristine ego. Thus, for Fairbairn libido

becomes the energy of the primary life drive of the ego, and energy
and structure are not as separated as in Freuds conception of the id
and ego. I mention the Fairbairn or Guntrip approach primarily to
reject it, because it is clear that if one accepts this terminology then
one must totally reject the entire metapsychology of Freud, which I
choose not to do. In addition to this, the Fairbairn-Guntrip approach

contains tremendous terminological confusion which in no way can
be resolved by simply further study but has within it a worse inherent
inconsistency than the standard metapsychology.
The commonly accepted concept of the ego as presented by
Freud (1923) and refined by Hartmann and many others finds the ego
as a substructure of personality defined by its functions. More

specifically, these days we may think of four basic sets of apparatus
that define the ego functions. First, of course, are all the defensive
functions of the ego made famous by Freud and Anna Freud.
Hartmann (1950) introduced the conflict-free sphere of the
ego. This contains the primary autonomous functions, such as
memory, judgment, thought and reality testing, and the secondary

autonomous functions.
These secondary autonomous functions, such as orderliness,

are defined as functions which at one time were drive-connected or in
the service of defense against unacceptable impulses but now during
the course of epigenesis have developed a usefulness of their own in

adaptation and therefore become part of the conflict-free sphere of
the ego. Thus through a change of function, (Hartmann 1950) what
started in a situation of conflict may secondarily become part of the

nonconflictual sphere and come to serve different functions, like
adjustment, organization, and so on.
Furthermore, beside defensive functions, primary autonomous
functions and secondary autonomous functions, there are in the ego
what we call microinternalizations. These microinternalizations are
adaptative techniques and regulative techniques which the infant and
child learns by observing the parents and significant others around

him. Gradually in optimal circumstances he takes on these techniques
of adaptation in an ego-syntonic way, so these become a part of the
individuals concept of himself and a smooth part of ego functioning.
To further illustrate what is meant by these

microinternalizations, we can contrast them with
macrointernalizations, sometimes known as introjects. Introjects are
foreign bodies within the ego. They are due to intense
incorporation, which takes place when there is a phase-inappropriate
disappointment or the loss of a significant individual in the persons
life. These macrointernalizations, or introjects, have in a sense a
psychic life of their own within the ego that carries on a dialogue with
the individual, speaking to him and influencing his life. Schafer (1968)
presents a detailed discussion of introjects.
Finally, within the ego we find self-representations and object

representations, which will be discussed in detail later. The important
point to be made on this subject with respect to the ego is that
original and early self-representations and object representations
gradually become attuned to reality and modified by reality under

optimal conditions and so become smoothly diffused in the
functioning ego systems. Thus when the self-representations and
object representations are correctly and realistically integrated into

the ego, the individual is able to use realistic methods of adaptation
that is to say, adaptation improves proportionately to our ability to
have more realistic and clear representations of ourselves and the
objects toward whom we have to adapt.
An object is a personreal, out there, repeatedly

experienced. When the significant objectsoriginally the mother of
courseare not too painful or frustrating, object representations are
fused into a realistic representation that one may adaptively relate to

and deal with. By the age of eight months, the first capacity for self-
object differentiation permits this process to begin. It is a process
which develops into what we call object constancy and eventually,
object love. A continuing interaction with significant objects,
especially the mother and later the father, siblings and peers, is
required for object constancy and object love to develop.
Notice the difference between object relations and object love.
It is possible to have many object relations; a good politician knows

the names and a little about many, many people. This is not the same
as object love, in which there is a genuine affection and empathic
feeling for other people as human beings with needs of their own.
Notice also that we speak of the epigenesis of the ego (Erikson 1959)
as an unraveling of inborn potential contingent on environmental
interaction. Thus the individual comes into life with certain anlage, or
primitive ego apparatuses, which unfold in their potential as a result
of successful interaction with those around him, and furthermore,
each stage of ego development depends on a successful experience
and development in the prior stage.
In addition to debate about the concept of ego, there is debate

about the concept of superego even within the ranks of classical
psychoanalysts. With the exception of the Kleinians, most authors
agree that the major phase of superego formation occurs during the
resolution of the Oedipus complex, around five or six years of age. Of

course, there are forerunners of the superego before that time. (The
Kleinians conceive of the main structure of the superego as forming
much earlier.)
There is a harsh critical aspect of the superego. The

forerunners of the harsh critical superego are memory traces of
punishment. The harsh critical superego is supercharged with
internal aggression; there is a psychic connection between the id and
the harsh critical superego that allows the draining of the aggressive
energies from the id via the harsh critical superego. Thus we see
people who are flooded with profound aggression turning this

aggression upon themselves through the harsh critical superego.
A benign aspect of the superego also exists, based on memory
traces of love and approval from the parents. The final precipitation
of the harsh and benign aspects of the superego occurs under the
influence of the Oedipus complex, and at this time there occurs the
major internalization of parental values.
The ego-ideal aspect of the superego is less clearly understood.
It seems to bridge the ego and the superego and to form out of
microinternalizations, which I have discussed already. Freud often
uses the term ego ideal and superego interchangeably, and he means,
by either, the internal values of the parentsgoals and identifications
that are formed in the interaction with the parents. Later authors
have used the term ego ideal to represent a separate aspect of the
superego. The traditional view was to see the ego ideal primarily as
an aspect of the superego and as representing something that the
individual would like to be. Thus, for example, shame is thought of as
a tension between the ego and the ego-ideal aspect of the superego in
which the individual has not lived up to his ideals; guilt is experienced
as a tension between the ego and the harsh critical superego in which
one has transgressed the rules internalized from the parents.
Later thinking such as of Kohut (1971) sees the ego ideal as
coming from the search for an idealized parent imago. When the
superego is formed the idealized parent imago is then placed in the
superego. However, both the superego and the ego become infused
with love and admiration for the idealized powerful parent. The
idealized parent becomes an aspect toward which one aspires and in
that sense becomes part of the ego ideal. We speak of the ego ideal as
pulling the ego forward. Thus, for Kohut the ego ideal is not a separate
functioning entity in the superego, but a kind of infusion of the
superego with power and admiration and consequently with a wish to
be like and to obey the beloved, internalized, idealized parent or

superego.
The term transitional object was first introduced by Winnicott
and later referred to by Modell (1963). For our purposes, we use
Modells definition of a transitional object phase of the development

of object love, during which there is a clinging, dependent
relationship to the object. This stands between the primary
narcissism, where there is no recognition of the object as separate,
and true object love, where there is the capacity to relate to the object
as separate, human and having needs of its own.
The concept of self-object was introduced by Kohut (1971) to
help distinguish between object relations and object love. The small
child has object relations but not object love. It relates to others as
selfobjects in which the object is experienced as part of the self and
having no life of its own. These are very important concepts, because
they appear in the transferences of many patients and they help the
therapist to understand certain aspects of behavior in psychotherapy
that ordinarily would cause irritation and rejection on his part. If one
understands a clinging, dependent transference in terms of the phase
of object love that the patient is in at the time, or if one understands
that the rage of a patient when he has to leave at the end of a session
or when the therapist goes on vacation represents the total inability
of the patient to think of the needs of the therapist or to tolerate any
lack of control over the therapist, then a more appropriate response

and interpretation can be presented to the patient.
The term disavowal was introduced by Freud late in his life to
describe the splitting of the ego in the service of defense. The most
recent description of this phenomenon in the literature is the so-
called vertical split of Kohut (1971), involving grandiosity. In
disavowal, both aspects of the ego are more or less conscious, but one
part is really unaware of or glosses over the other. The important

point is that repression is not involved. Disavowal is an important
mechanism of defense that occurs before the establishment of the

repression barrier and it stands along with such defenses as denial or
hallucinatory omnipotence. There is disagreement whether disavowal
ought to be called a splitting of the ego or a splitting of the self in the

service of defense. The important point is that two contradictory
perceptual or behavior systems are operating at the same time and
are both conscious; an individual shifts back and forth between them.
Freuds well known example was the denial of castration that the
little boy experiences when he first sees the female genital; the
perception that there are some humans who have no penises is kept
in the conscious right along with the insistence that all human beings
have penises. Both perceptions are reacted to and contained in the
conscious mind at the same time.
Kohuts vertical split is somewhat different and has to do

with the conscious presence of a grandiose self in the mentation and
behavior of an individual who is also aware at the very same time that
he is not grand. For example, the patient at times consciously behaves

as if he were powerful and omnipotent and important in a manner far
out of proportion to his real state; at other times he behaves
appropriately but in a manner contaminated by low self-esteem,
shame propensity and hypochondria.
The repression barrier is established firmly at the time of

resolution of the Oedipus complex. Transference represents a
crossing of the repression barrier in which an object representation
or a self-representation is projected onto the therapist and he is then

related to accordingly. This is the restricted definition of transference;
there are others. Therapeutic alliance, on the other hand, is an
attitude of expectation and cooperation based primarily on memory

traces of previous experience with authority figures or doctors. It has
nothing to do with the repression barrier, no crossing of which is
involved. It is based primarily on memory traces of successful
experiences with people toward whom the individual had turned for
help in the past.
Clearly the style of defenses as well as the primary

autonomous functions of the ego involved have a genetic or
hereditary basis. To put it technically, Nagera (1967) speaks in terms
of ego structures rather than ego apparatuses or ego functions, a view
which postulates the existence from the very beginning of life of a
number of ready-made primitive structures or organizations and
which is completely in line with Hartmanns assumptions of inborn
ego apparatuses belonging to the conflict-free sphere. These

structures exist at birth, while most other structures have to be
created during the development; that is, further structuralization is
taking place all the time as development proceeds.
Three factors are crucial as determinants of the rate and extent
of acquired structuralization. First we have innate limitations, which
differ from individual to individualdifferences in inborn qualities.
Second, human needs and human nature are involved; the various
degrees and intensities of our needs serve as a triggering force to
propel us along the path of structuralization and ego development.
The third factor is the environment in which we happen to be born.

Adaptation to a high degree of civilization also demands a great
extent of ego structuralization. As civilization becomes more complex,
it makes further demands on the egos capabilities to deal with the
ever-increasing complexity of propositions of the new order of things.
Thus, as Nagera points out, education is basically a system

devised to teach children in a condensed and simplified manner the
means by which they can build complex psychological ego
apparatuses capable of dealing with the complexities created in our

world. All education does is to exercise a number of mental
capabilities in special directions and combinations until the ego
learns to perform a number of complicated functions in interaction.
I suspect that the capacity of the borderline to snap back and
to move up and down the ego axis is related to these inherited styles
of defense and primary autonomous functions as well as to the

existence of at least some adaptive identifications with some
significant parental or grandparental figure in the past. Sometimes

these identifications dont occur until adolescence, during which the
borderline patient is lucky enough to meet a relatively healthy
individual that he can use for microidentifications.
Rapaport (1951) explains that the ego is born out of conflict
and is a party to it. Certain apparatuses of the ego have primary
autonomy and certain functions are outside of conflict in the conflict-

free sphere of the ego. Now there is a certain constancy and reliability
in the autonomy of the ego as an emergent organization which has
laws of its own distinct from the elements which control it. Rapaport
warns us that this autonomy, especially the secondary autonomy, is

always relative. He indicates that the onslaught of drive motivations,
especially when unchecked by therapeutic help or when aided by

overzealous therapeutic moves, may reverse the autonomy and bring
about a regressed psychotic state in which the patient is to a far-
reaching extent at the mercy of the drive impulses. The higher-level

autonomous motivations are dissolved and the allies that the
therapist usually counts onspontaneity and synthesisare absent.
He points out, Thus we can see that the issue of ego autonomy is not
merely a theoretical problem, but also a practical one of therapy,
particularly in borderline and psychotic cases.
He doesnt say it, but the converse is obviously also true. Good
psychotherapy enhances the core of autonomous ego function, leads
to greater cohesiveness of the selfbetter overall ego functionand
reduces the clamor and disturbances of the impulses which is always
such a big problem in treatment of borderline patients.
Gitelson (1963) points out that it is very difficult to separate
what is constitutionally given and what is acquired through
introjection. The climate of the first relationship with the mother

establishes the fundamental and typical mood of the person. The
intimacy of the symbiosis of mother and child is what makes it so
extremely difficult to be certain what is constitutionally given and

what is acquired. Of course, everything is further complicated by
subsequent identification with father and siblings and other
significant persons later on. As a matter of fact, the latter, including
the parents as they are and appear in later years, impose various
modifications for better or for worse, on the original identifications.
However, Gitelsons point is that these original identifications

are essentially indestructible and retain their effectiveness. The earliest
identifications enter into the formation of what we call ego nuclei, and
a great deal in the therapy depends on the extent to which these are
dissolvable and can be dealt with by therapeutic modification. It is not
difficult to see, then, how complex the careful dissection of an
individuals ego functions can be and how difficult it is to understand

the basic nuclei of the ego in the borderline patient. Now we can also
see that these ego nuclei, the earliest identifications, vitally affect
what we call ego strength.
The understanding of so-called ego strength also depends on
understanding of the superego. The internalization of the parents via
identifications, which reaches its peak at the Oedipal stage,

internalizes both critical and loving aspects of the parents. Thus there
is both a harsh or critical superego and a benign superego. Obviously

inner strength is based on the presence of the benign superego, which
infuses the ego with approval. Day-to-day parental approval
generates a whole system that causes the build-up of ego skills in the
latency period. If the parent withdraws at that time, the harsh or
critical superego takes over this function, and the unfortunate
individual is quite vulnerable. I have noted that in adolescence the

more malevolent of peers are often allowed to take over this function
also.
The ego ideal is also in the superego and is related by some
authors to the equivalent of the primary narcissism of the baby. Thus,

one tries to live up to the idealized image of what one could be, and
this is a part of the self-regulation system. For example, let us take

ambition. From the point of view of libidinal drives, ambition can be
understood as the wish on the oral level to incorporate the world, on
the anal level to produce the largest bowel movementto be

productiveand on the phallic level to have the biggest capacity, to
be outstanding and efficient. Simultaneously, ambition can be
understood on the level of the ego ideal or narcissism as the wish to

be magical, powerful in sublimated socially acceptable ways, and so to
get love and approval from the self and others.
Therefore, if one has a good ego-ideal system, then one can
mobilize ones ego potential. Otherwise one must search outside
oneself for an idealized parent imago to be with and must use ones
conception of what the leader (the idealized parental image) is in
order to regulate oneself. This is a more technical description of
Odiers neurosis of abandonment described in borderline cases in
Chapter 2.
The secret behind ego strength, then, from the point of view of
the superego lies in the presence of a benign, approving, self-
regulating system made up of the expectations one has of oneself
which come from the ego ideal and the approval of oneself coming
from the benign superego.
Gitelson (1963) further explains that the primary influence of

the parent is internalized and made permanent in the formation of
the superego. He writes, Aggressive and libidinal energy which is
withdrawn from its original focus on the parents becomes available to

the superego itself, providing it with powers which are exerted
against the forces of the id and the derivatives in the ego. The
superego originates in part from the conditionings which occur in the
preverbal and pregenital relationship to the mother; it is transiently

stabilized in the context of the Oedipus complex, in middle childhood;
it appears in its ultimate form after puberty, and in this form it is
crucial to the definitive molding of character. However, its strength is

relative not only to the quality of identifications which have entered
into its formation, but also to the strength of the id.
We see, therefore, that the metapsychological description of
the borderline patient is extremely complicated; in our explanation of
the symptomatology of the patient we have to take into account both
the ego and the superego. More precisely we must study the four
functioning subsystems of the egothe primary autonomous

functions, the secondary autonomous functions, the defensive
functions and the adaptative identificationsAND the subsystems of
the superegothe harsh critical superego, the benign superego and

the ego ideal. The interaction of these subsystems produces the self-
regulating system which the patient presents clinically and upon
which he must rely in dealing with the vicissitudes of life.
Since there are a number of subsystems that can go wrong, we

are dealing with a tremendous variety of possible permutations and
combinations that produce a variety of metapsychological
characterizations for the borderline patient. Thus, there is no reason

to believe that any one metapsychological characterization can be
applied to all borderline patients; it is more scientific to begin with a
given borderline patient at the descriptive level of ego and superego
functioning and then try to understand the nature of the various

subsystems described above and the intrasystemic defects and
conflicts that have led to symptomatology manifested by the specific
patient.
Kernberg (1970b) proposes a classification of character

pathology on the basis of careful examination of instinctual
development, of superego development, of defensive operations of
the ego, and of the nature of the pathological character traits as well
as of the vicissitudes of internalized object relationships, and he
outlines the structural characteristics of higher, intermediate, and
lower levels of organization or character pathology. I wont take the
space to review this paper in detail, but it gives the reader an idea of
the number of permutations and combinations that take place, many
of which fall under the general rubric of borderline patient. Clearly
these combinations can fall at various levels on what one might call a
continuum ranging from the well integrated to the almost

disintegrated. In a way it is the existence of this continuum that has
caused so much trouble, because there are no discrete diagnostic
entities here, but only more or less of this or that, and it is difficult at
times to be certain that two authors who are talking about
borderline patients are really talking about the same classification
of patients on the continuum.
I think this accounts for a great deal of the profound

disagreement in the literature. For example, Masterson (1972) has
devoted a whole book to the basic theory that separation from the
mother does not evolve as a normal developmental experience for the
borderline patient. On the contrary, it entails such intense feelings of
abandonment that it is experienced as a rendezvous with death. To

defend against these feelings, the borderline patient clings to the
maternal figure, thus failing to progress through the normal
developmental stages of separation and individuation to autonomy
(Masterson and Rinsley 1975). This is of course based on the concept

of separation-individuation, which has evolved from Mahler (Mahler
and Gosslinger 1955; Mahler and LaPerrier 1965; Mahler et al. 1975;
Mahler 1963, 1971, 1974, 1975), who studied by direct observation

the separation-individuation processes of normal children. It is based,
therefore, on certain fundamental assumptions about the phases that
normal children go through (the autistic phase, the symbiotic phase

and the separation-individuation phases).
Great theoretical differences appear at once if one looks at the
work of Kohut (1971). Mahlers work is based on a theoretical system

defined by the position of an observer equally distant from the
interacting parties of mother and child, occupying an imaginary point
outside the experiencing individual. Kohut insists that the core area of
psychoanalytic metapsychology is defined by the position of an

observer who occupies an imaginary point inside the psychic
organization of the individual, with whose introspection he
empathically identifies. This is of course accomplished through the

transference revival of childhood experience rather than through
direct observation of children and through reconstructions of the
inner life of the child on the basis of transference reactivations. This

difference of viewpoint between Mahler and Kohut points to why
there is such a profound theoretical difference between those who
work from the classical psychoanalytic point of view and those who
approach the borderline patient from a variety of other types of

viewpoints such as interpersonal theory or social interaction or
behavior theory and so forth.
The crucial differentiation among (1) narcissistically
experienced archaic self-objects; (2) internalized psychological
structures that perform drive-regulating, integrating and adaptative

functions previously performed by external objects; and (3) true
objects cathected with object-instinctual investments forms the
foundation for recognizing the profound distinction between
psychoanalytic metapsychology and all other points of view (Kohut,
1971).
CHAPTER 6
Metapsychology of Narcissism
The greatest current debate, of course, is on the subject of

narcissism. We can distinguish two basic views about primary
narcissism. Balint (1968) flatly states that there is no such thing as

primary narcissism. In his view, the individual is born with primary
object lovethe kind of seeking is for an object that will gratify the
person without his even having to first communicate the need to the
object. It is in the wish for the intuitive, empathic all-loving maternal
object. There is no room in the theory of Balint for primary

narcissism, and he sees development as progressing strictly along the
line of object relations, from primary object relations to mature object
love.
Freud, on the other hand, beginning with his famous paper On
Narcissism (1914), defined narcissism as the cathexis of the self or
the ego with libido. Notice that Freud sometimes uses the terms ego
and self interchangeably. There is only a dim notion in his theoretical
formulations that a distinction is necessary between self and ego. This
is usually called the syphon theory of narcissism, because it is

thought of as a U-tube in which there is a fixed quantity of libido
available. When more is cathected to the ego or the self, less is
available to be cathected to objects, and vice versa. According to

Freuds theory, the infant passes from a phase of autoerotism, in
which there are simply body states before any ego nuclei have even
developed, to a phase of primary narcissism, which begins with the
formation of ego nuclei and represents an overwhelming cathexis of
these ego nuclei with the libido. Then occurs a gradual transition from
the stage of primary narcissism to the state of object love, as libido is
divested from the ego and cathected to objects (object-
representations).
Secondary narcissism is also defined differently by different
authors. Freud defined secondary narcissism as a defensive
withdrawal of libido from objects back into a cathexis of the ego or

self, and that was the usually accepted definition of secondary
narcissism until recently. Balint insists that all narcissism when it
appears is secondary narcissism, since there is no such thing as
primary narcissism. Disregarding the concept of secondary

narcissism, Kohut, at the other extreme, would argue that narcissism
follows an independent line of development and reaches levels of

narcissism from primitive to mature.
In his theoretical orientation Kohut, in direct contrast to Balint
(1968), stays closer to the classical psychoanalytic viewpoint,

postulating a preliminary phase of autoerotism followed by primary
or primitive narcissism. Kohut (1971) explains that in normal
development, The equilibrium of primary narcissism is disturbed by

unavoidable shortcomings of maternal care, but the child replaces the
previous perfection (a) by establishing a grandiose and exhibitionistic

image of the self, the grandiose self, and (b) by giving over the
previous perfection to an admired, omnipotent (transitional) self
object: the idealized parent imago." With much greater care and
attention than any author has previously paid to this subject, Kohut
elaborates and distinguishes the vicissitudes that occur as the
equilibrium of primary narcissism is inevitably disturbed.
Ferenczi (1950) as early as 1913 points out that an analysis
that reaches to the depths reveals that feelings of inferiority are
reactions to the exaggerated feelings of omnipotence to which certain

patients have become fixed in their early childhood and which have
made it impossible for them to adjust themselves to any subsequent
renunciation. The manifest seeking for greatness that these people

have, Ferenczi explains, is the return of the repressed, a hopeless
attempt to reach once more, by means of changing the outer world,
the omnipotence that originally was enjoyed without effort. Those
objects which are not as yet available to the immature psyche will be
experienced in an intrapsychic world as if they were parts of the self.
Freud would say these need-fulfilling objects are invested with

narcissistic libido. Kohut feels that the line of development of object
love proper can only begin after the secure differentiation of the self
from objects. Before that time objects are not loved for their
attributes, which are at best dimly recognized, they are loved as part
of the self.
Modells (1968) theory, which I have referred to previously,

states this in terms of the child continuing to need to create illusory
substitutes that he can control in place of the actual mother, who has
an independent center of volition or will power. Modell, following
Winnicott, calls this a transitional object relation, and he explains how
by utilizing fantasies of omnipotence the infant is enabled to preserve
in one aspect of his mind his illusion of symbiosis.
Now whether these archaic self-objects are referred to as
transitional objects or part objects or self-objects seems to be mainly

a matter of which author you read. The point is that these various
object imagos are gradually sorted out realistically and, concurrent
with the achievement of stable reality testing, they are consolidated

into whole objects with stable characteristics. This opens the way to
true object love.
Notice that this discussion of object relations is focused on the

intrapsychic world and avoids a confusion of the early phases of
development with later stages. The failure to make this distinction

has been responsible for controversy and has been the source of one
of the primary difficulties between the classical psychoanalytic
approach and the attempts of Melanie Klein and Fairbairn and others.
Modell is criticized by Kohut and his co-workers (Gedo and Goldberg
1973) for failing to attain the necessary metapsychological clarity in
his view that cognitive differentiation of self from object marks the
emergence of the self as a cohesive entity. By differentiating self-

objects from those invested with true object libido, Kohut presents an
evolving sequence of the childs objects and also the complementary
issue of the development of the childs self. Thus the central clinical
discovery here is the overriding importance of the attainment of a
cohesive sense of self Clearly, failure to achieve this cohesion is

characteristic of various forms of severe psychopathology, and even
in those cases that Kohut feels are analyzable there is a vulnerability
to fragmentation of the self under stress.
Kohut believes that there is a separate normal developmental
line for narcissism, whereas the other authors believe narcissism
should essentially disappear and change into object love in normal

development. For Kohut, then, after primary narcissism has been
established, the next developmental stage is an attempt to retain
primary narcissistic equilibrium in the face of realistic disappointing

experiences. This stimulates two psychic formations, the grandiose
self and the idealized parent imago. The child forms a fantasy of being
himself as all-powerful and omnipotent, a sense of himself as

grandiose. He also forms an imago of an all-powerful omnipotent
idealized parent. Both of these formations occur around the same
time and represent the attempt to retain primary narcissism either by
imagined grandiosity or by connection with the omnipotent parent

who will therefore meet all his needs.
The grandiose self and idealized parent imago are normally

eventually integrated into the personality. The idealized parent
imago, as already mentioned, becomes part of the ego ideal in the
superego toward which the individual strives. The grandiose self

becomes a part of the ego apparatuses and functions as ambition and
drive regulation. Kohut explains that the grandiose self, when it is
integrated into the ego, pushes the individual forward through

ambition and drive regulation. The ego ideal, formed of the idealized
parent imago, pulls the individual from above, so to speak, and

becomes a goal toward which the individual strives.
A cohesive sense of the self as separate from others must occur
in normal development, and it is based on the reality fact that one is
separate as a mental and physical self. The cathexis of various self-

representations determines how we think about and even behave
towards our real mental and physical self.
The term identity is probably the most confusing of all and has
been made even more confused by Eriksons use of the term ego-
identity. Identity is best thought of as a sociological term, referring to

the individual as a coherent entity with direction and continuity at
any given stage of human development. Identity is a vector term; thus
it has both a magnitude and a direction in contrast to all these other

terms, which are essentially scalar (only magnitude). There must also
exist identity representations. The subjective feeling of having an
identity that Erikson emphasizes would depend primarily on the
cathexis of identity representations. The concept of identity is most

complex because it depends on all the other concepts already defined
as well as on the vicissitudes of the milieu and the culture in which

one lives. For example, when we speak of sexual identity, we are
really including a host of other functioning aspects of the personality
as well as interpersonal interactions and the influence of the social

milieu, all of which form representations in the persons ego of his
sexual identity.
Kohut explains how under ordinary circumstances, the
grandiose self and the idealized parent imagoimportant for

ambition, enjoyment and self-esteembecome integrated into the
adult personality. With each of the mothers inevitable minor
empathic failures, misunderstandings and delays, the infant
withdraws narcissistic libido from the archaic imago of unconditional
perfection (primary narcissism) and acquires in its stead a particle of
inner psychological structure that takes over the mothers functions
in the service of the maintenance of narcissistic equilibrium. Thus,
tolerable disappointments in the primary narcissistic equilibrium
lead to the establishment of internal structures that provide the

ability for self-soothing and the acquisition of basic tension tolerance.
However, if severe narcissistic traumas are suffered by the child, then
the grandiose self and the idealized parent imago are retained in
unaltered form, not transformed into the adult personality, and exert
a pressure of their own.
Another way to understand the lack of inner sustainment (Saul
1970) in the borderline patient is to focus microscopically on the ego

ideal. The core of narcissistic omnipotence in the ego ideal is the
remnant of the positive, gratifying, mother-child relationship when

there has in fact been good mothering. When this component is
present it has a need-satisfying wish-fulfilling quality of its own,
which continually bathes the ego with its own sense of narcissistic
omnipotence, no matter what other conflicts there may be. This
component of the ego ideal is either there or it is not thereand it
has a profound influence on the further growth and development of
other psychic structures.
Weiner (1973) explains, If the mothering is poor or

inconsistent and the substructures fail to develop, the resultant
inability of the infant to deal with a feeling of abandonment and
helplessness will distort ego growth and later object relations. Basic
trust cannot grow and fear of abandonment and betrayal beclouds all
later object relations. Internally, the ego cannot fall back on its own
psychic friends in times of trouble, for on this level, there are none. At
this time a feeling of hopelessness occurs, and this psychic state sets
the stage for the disposition to fall ill of melancholia (Freud, 1917).
Weiner (1973) also calls attention to various rescue

operations for narcissism described by Hartmann and Lowenstein
(1962). These rescue operations are very important. They involve the
idealization of positive aspects of the parents and other important

people. In addition the ego may develop character traits which fulfill
precepts of the ego ideal. The person may go through life having
developed character traits that satisfy the ego ideal, but the loss of an
important object or a breakup of these traits due to retirement,

financial reverses and so on may occur, taking away the external
narcissistic input, and then the person suffers a depression. At this

point the patient feels mentally bankrupt and external reassurance
becomes useless because the poverty about which the patient
complains is psychologically true. . . . Endless protestations from the

family, friends, and well-meaning professionals are usually addressed
to the conscious ego, which is weakened, while the unconscious ego is
preoccupied in a fruitless search for inner nurturance. Many
borderline patients come for psychotherapy at this point.
Kohut describes the difficulties between the child and the

parent that lead to disruption of the normal trend of events in the
vicissitudes of narcissism. Under optimal circumstances the child
experiences gradual disappointments in the parent or idealized
object; to put it another way, the childs evaluation of the idealized
object becomes increasingly realistic, which leads to a withdrawal of
the narcissistic cathexes from the imago of the idealized self-object
and to their gradual internalization.
If the child suffers traumatic loss of the idealized object or
phase-inappropriate disappointment in it, then optimal

internalization does not take place, and the psyche remains fixated on
an archaic selfobject:
The personality throughout life will be dependent on

certain objects in what seems to be an intense form
of object hunger. The intensity of the search for and
of the dependency on these objects is due to the fact
that they are striven for as a substitute for the
missing segments of the psychic structure. They are
not objects .. . since they are not loved or admired for
their attributes, and the actual features of their
personalities, and their actions are only dimly
recognized. They are needed to replace the functions
of a segment of the mental apparatus which has not
been established in childhood.
The trauma suffered most repeatedly in these cases is severe
disappointment in the mother, who because of her defective empathy

with the childs needs did not appropriately fulfill her functions as a
stimulus barrier, an optimal provider of needed stimuli, a supplier of
tension-relieving gratification and so on, depriving the child of the

gradual internalization of early experiences of being optimally
soothed or aided in going to sleep. The mature psychic apparatus
should later be able to perform these functions predominantly on its
own.
This has been carried to a fascinating degree of further
metapsychological refinement by Kohut. He writes, as noted above, of

phase-inappropriate disappointment in the idealized parent imago:
depending on the phase in which this inappropriate disappointment
occurs we will see different clinical pathology. Thus in the very early
stages of life with the predominant need for maximal soothing and
relaxations towards going to sleep from the parent, a phase-
inappropriate disappointment will lead to the search for this kind of
soothing from the outside, since the proper transmuting

internalizations have not built it in from the mother. The patient is
left with a malfunctioning stimulus barrier, and we observe the
search for drugs or other procedures (including psychotherapy) for

the primary purpose of obtaining this soothing from an external
source.
Disappointment in the late preoedipal period leads to a
sexualization of pregenital drives and derivatives as development

proceeds, with a resulting predominance in the psychic life of
perversions of all sorts in fantasy or even acting out.
Disappointment in the Oedipal or early latency period makes the
internalization of the idealized parent imago into the ego ideal

impossible (or, in early latency, undoes this internalization, which at
this point is new and shaky) and results in a fixation on the search for
an idealized parent outside of the patient. The result is an intense
striving for a dependency on an idealized person, as if this person
were a missing part of the self; approval from this person is required
to maintain narcissistic equilibrium, and personal accomplishment
brings no lasting satisfaction. Odiers neuroses of abandonment fall
in this area.
As an example of the latter, Kohut presents a patient in which
the central defect of his personality was the insufficient idealization
of his superego (an insufficient cathexis with idealizing libido of the

values, standards, and function of his superego) and, concomitantly,
the strong cathexis of an externally experienced idealized parent
imago in the late pre-Oedipal and Oedipal stages. This led to a diffuse
narcissistic vulnerability, the hypercathexis of his grandiose self
occurring mainly in response to disappointments in the idealized
parent imago, and the tendency toward the sexualization of the
narcissistically cathected constellations. In such patients a

hypersensitivity to disturbances in the narcissistic equilibrium takes
place with a tendency to react to sources of narcissistic disturbance

by a mixture of wholesale withdrawal and unforgiving rage, forming a
very typical and frequent clinical picture.
According to Kohut, an unconscious attachment to, and failure

in integration of, the archaic grandiose self or the idealized parent
imago and their corresponding self-object representatives, result
from impaired development of narcissism. Disturbance in integration

of the grandiose self means that the primitive narcissistic impulses
remain walled off from the reality ego, but they continue to influence
the self, as manifested by wide oscillations in self-esteem and other
phenomena.
A rather strongly opposing view is presented by Kernberg
(1970a, 1974a, 1974b, 1975a, 1975b), who refuses to differentiate
between narcissistic personalities and borderline personalities in the

way that Kohut does. Their fundamental similarity, he insists, is in the
predominance of mechanisms of splitting or primitive disassociation,

as already briefly described in Chapter 3. Thus, from a dynamic
viewpoint pathological condensation of genital and pregenital needs
under the overriding influence of pregenital (especially oral)

aggression is characteristic of narcissistic personalities as well as
borderline personality organization.
The difference between a narcissistic personality structure and

the borderline personality disorder, says Kernberg, centers on the
specific presence in the former of (1) an integrated grandiose self
which reflects a pathological condensation of some aspects of the real
selffor example, the specialness of the child that may have been
reinforced by the parents; and (2) the ideal selfwith fantasies and
images of power, wealth, and beauty that compensated the small child
for the experience of severe frustration, rage and envyand the ideal
objectthe fantasy of an ever-loving and ever-giving mother in
contrast to reality.
Kernberg argues that the integration of this pathological

grandiose self compensates for the lack of integration of the normal
self-concept and explains the paradox of relatively good ego
functioning and surface adaptation in the presence of a predominance

of splitting mechanisms and lack of integration of object
representations.
The disagreement is about the origin of this grandiose self and
whether it reflects the fixation of an archaic normal primitive self
with a separate line of development (Kohut) or is always a

pathological structure clearly different from the normal infantile
narcissism. Kernberg emphasizes the pathology, and especially the

repetitive chronic activation of intensive rage reactions that comes up
in the psychotherapy of the borderline patient.
CHAPTER 7
Disputes and Disagreements
Theories of Melanie Klein.
Jacobson (1964) made what might be called the first heroic
attempt from the point of view of classical ego psychology and
metapsychology to make sense out of Melanie Kleins confused
concepts of ego, self and identity. She evaluated Kleins contributions

and pointed out that there was a failure in her work to distinguish the
endopsychic representation of external objects from introjects and
from the infantile superego. In other words this differentiation, which

I have mentioned previously, between archaic internalized objects,
objects out there and representations of objects, is not carefully made.
Jacobson points out that Klein fails to distinguish the constitution of

self, and object representations, object relations, and ego
identifications from superego formation. Jacobson makes a very

careful attempt to describe the vicissitudes of self-and object
representations as they enter both ego and superego formation.
An attempt to sharpen up the theoretical approach of Klein has
been made by Fairbairn, Guntrip and others; these authors make an

effort to bridge the distance between the Kleinian approach and the
classical metapsychological approach. Fairbairns theoretical
contributions suffer from a confusion of concepts similar to Kleins. In
addition, we have the criticism of Klein from the point of view of
Winnicott and Balint, who theoretically are closer to Guntrip and
Fairbairn but clinically point out that Kleins tendency to push farther
and farther back the age at which mental mechanisms appear tends to
neglect the influence of the environment.
Thus the vagueness and ambiguity of Kleinian terminology are
major stumbling blocks preventing the clarification of Kleinian theory

even by such brilliant minds as Guntrip and Fairbairn. There are
internal inconsistencies within Kleinian theory, and because of them I
am not going to spend much time discussing the Kleinian school and
their approach to the borderline patient. In addition, there are shifts
in the way the terms ego and self are used throughout Kleinian
literature that make it very difficult to grasp the exact meaning of
such important concepts as projective identification. Every time I

give a course on the borderline patient this concept comes up, and it
is very difficult to pin down just what Klein meant by it. She originally
described it as the projection of split-off parts of the self into another
person. One aim of the process, then, is the forceful entry into the
object and control of the object by parts of the self. At this point she
uses the concepts of ego and self interchangeably, whereas elsewhere
she describes them separately. Her followers have broadened the
concept of projective identification in a variety of ways that I wont go
into at this point. We will discuss projective identification later in

more detail (see Chapter 9).
Theories of Primitive Object Relations.
The theory of primitive object relationships (see review by L.
Friedman 1975) is the basic point of disagreement, and yet it is vital
to any understanding of the borderline patient. A reasonable and

relatively simple theory of primitive object relations and the
application of this theory to the understanding of the borderline

patient is presented by Modell (1963). Modell insists that the
borderline patient is a homogeneous group, different than the
schizophrenic patient on the one handwhere there is a significant

disorder in sense of reality testingand from the neurotic on the
other hand. This is manifested by the primitive but consistent form of
object relationship that the borderline patient forms in the

transference.
To my knowledge this is the first serious attempt to
differentiate the borderline patient on the basis of a consistent type of
transference that he forms. Modell reminds us that he is using the

term borderline as I havenot as ambulatory schizophrenia but as a
separate group. The object relationship that the borderline patient

forms in the transference, according to Modell, is midway between
the transference of the neurotic and that of the psychotic. In the
neurotic transference, the object is perceived as outside the self and

invested with qualities that are distorted fantasies arising from the
subject. But the object still exists as a separate individual. On the
other hand, schizophrenics are unable to perceive that there is

something outside of the self at all. Therefore, Modell sees the
borderline transference as related to a transitional stage, and he
compares the relationship of the borderline patient to his physician as

analogous to that of a child to a transitional object, the blanket or a
teddy bear. He states, We can observe that there is a uniform, almost
monotonous, regularity to the transference phantasies, especially in

the opening phases of treatment. The therapist is perceived invariably
as one endowed with magical omnipotent qualities, who will, merely
by his contact with the patient, effect a cure without the necessity of
the patient himself to be active and responsible.
The type of transference formed by the borderline, according
to Modell, is also transitional because it stands midway between a
state of affairs where there is an absence of the sense of self, as in the

psychotic, and one where there is a distinct sense of self, as in the
neurotic. There is a fusion or confusion of the sense of self with the
object.
It is implied from this that there are three phases of object love
that take place in early development. In the earliest phase the young
infant responds to the mother but cannot make a psychological
distinction between the self and the object. The middle phase is the
phase of transitional object relations. Then mature object love,
according to Modell, is the stage where there is a distinct separation
between self and object. Thus the difference between the borderline
and the neurotic patient resides in the fact that for the most part the
psychic development of the neurotic patient has passed through the
stage of the transitional object, whereas the psychic development of
the borderline patient became arrested at the stage of the transitional
object.
The cause of this arrest essentially has to do not with physical
loss of the mother but with the kind of failure of mothering that I have
already discussed in Chapter 2. Modell similarly describes from his
clinical experience, not mothers who are lost, but mothers who are
unable to make emotional contact with their children as they

themselves are severely depressed or even psychotic. There is a
significant comment that in some of these cases the usual amount of
holding and cuddling was absent. In other patients, although the

physical care was adequate, there was a profound distortion in the
mothers attitude toward the child. For example, the mother could not
perceive the child as a separate person, which induced a relative

incapacity on the childs part to differentiate self from object. My
clinical experience has been very much of the same nature from
reconstruction in the intensive psychotherapy of the borderline
patient.
Why is it that the borderline group does not slip into
schizophrenia? Why do they hold on to the capacity of reality testing?
According to Modells theory, the schizophrenic patient is fixated at
an even earlier stage of development, but in addition to that Modell

postulates possible biological factors. There is no point in going into
this, because it is an X factor that one could speculate on endlessly.
Self and Identity.
Kohuts contribution, differentiating selfobjects from those
invested with true object libido, makes it possible for the first time to
meticulously clarify the overriding importance of the attainment of
the sense of the cohesion of self. It is this consolidation of self that
others, like Jacobson, have referred to as the stable sense of identity.
Failure to achieve cohesion of self characterizes various forms of
severe psychopathology. Regressive fragmentation of the sense of the
wholeness of the personality corresponds to what Freud (1927, 1938,
1940) called splitting of the ego in 1927. In some places he used the
word ego differently from the way the word ego is used in the
structural theory, since at the time disovowal (Freud, 1927) takes

place, the ego is not firmly established. Here is a confusion again;
Gedo and Goldberg (1973) point out that the correct terminology
should be splitting of the self. (We reject Jacobsons term identity

because her usage of identity instead of self is some kind of an attempt
to straddle the gap between two disciplinessocial and individual
psychologyand doesnt really belong to either.)
Erikson (1959) compounded this terminological muddle, and

it is quite a muddle, by introducing the term ego identity to designate
the ultimate maturation of the sense of self in adolescence. Therefore,

it is best for us to drop the term identity, a social sciences term, and
stay with the construct of self. Even self is very semantically difficult,
and there are great problems in trying to impose it on the tripartite

model of the mind (ego, id, and superego), but the cohesive sense of
self is a very important aspect to clinically consider in working with
borderline patients!
Adult Narcissism.
The question remains unanswered whether pathological

narcissism in the adult is a consequence of some sort of pathological
narcissistic infantile organization or whether it is merely a result of
interference with the normal evolution of infantile narcissism, leading
to developmental arrest (see reviews of the debate by Ornstein 1973,
1974a, 1974b). Kernberg (1974b) insists that pathological
narcissism is strikingly different from normal narcissism. In his view,
the grandiose self and the idealized parent imago represent
pathological developments, not developmental arrests, and there are

no substantive agreements with Kohut in this area (Ornstein
1974a).
Zetzel (1971, 1973) maintains that narcissism is a

predominant element in the character structure of borderline
patients; she believes, This behavior often serves to cover a deep
distrust and helps to defend against underlying paranoid traits which
are based on the projection of a rather primitive oral rage. Thus

Zetzel is in essential agreement with Kernberg, seeing the personality
organization of borderline patients as impulsive and infantile. She

attributes this to an underlying weakness of the structure and
organization of the ego. The egos overall capacity for neutralization
of instinctual drive derivatives is poor.
I would suggest that we try to stay as close as possible to the
practical situation through the scrutinizing of clinical materialbegin
with the clinical material and try to see which of these various
theoretical formulations fit the given clinical phenomena. This means
that the intelligent reader will have to juggle in his head these
conflicting theories and be prepared to try to fit the theories to the
clinical phenomena the way one fits scattered pieces together in a
jigsaw-puzzle opening. You scan and try out several pieces, one at a
time, to see which piece fits.
Projection and Introjection.
Kohuts theory attempts to get rid of the concepts of projection

and introjection in infancy altogether, as these are terms that have
been utilized with very insufficient precision in the literature
(Rapaport 1944). Thus the so-called projections of psychotics are

caused by failure to maintain the boundary of the self. The attribution
of a thought or feeling of ones own to another is simply due to the
lack of differentiation between the self and object.
A repression barrier is necessary before genuine projections
can take place. Gedo and Goldberg (1973) attempt to differentiate

what one might call less mature types of projections from more
mature projections as seen in the neuroses. The latter occur after a

reasonably firm repression barrier has been established. The
repression barrier is not solidly established until the Oedipal phase of
life has been worked through. These authors imply that since the
irreversible establishment of the ego is not expected to take place
until after the resolution of the Oedipus complex, the concepts of
projection and introjection before that time are at least
metapsychologically quite different, if not altogether incorrect. Thus it

is much easier, relatively speaking, to characterize the borderline
patient in a descriptive fashion than it is to understand the
metapsychology involved.
According to Zetzel (1973), the defensive processes of internal
splitting of the ego into good and bad parts is supported and
engineered through the interplay of introjection and projection.

Meissner (1974) describes the correlated aspects of introjective and
projective mechanisms and how the interplay of introjective and

projective mechanisms weaves a pattern of relatedness to the world
of objects and provides a fabric out of which each individual fashions
his own self-image. Out of this interplay comes the gradual emergence
of differentiation between the self and object, according to authors
Zetzel, Modell and Meissner, without postulating a second line of
development for narcissistic libido.
However, Zetzel (1973) continues, the excessive operation of
these primitive mechanisms in borderline patients prevents the ego
from achieving any meaningful integration of both self-and object
images which have been built up out of libidinal derivatives, with the
self and object images which have been built up out of aggressive
derivatives. She sees a progression of cycles of the projection of

aggression and the subsequent reintrojection of hostile and
destructive object and self images as central to the development of
both the psychotic and borderline personality organization. In

psychotic development, this process produces a regressive refusion of
self and object images with the loss of ego boundaries and self-object
differentiation. In the borderline however, the process does not reach

that level of regression, but rather brings about an intensification and
fixation of splitting processes.
In a complex metapsychological sentence Zetzel claims, Thus
splitting achieves an active separation of introjects of opposite quality
good as opposed to bad. The integration of such object derivatives
is one of the major ways in which aggression is neutralized and
detoxified. This is explained in terms of the fusion of libidinal and
aggressive instincts in normal development, which permits a
neutralization of aggression. The splitting mechanism on the other
hand, results in an inadequate neutralization of instinctual energies;

according to Zetzel, Splitting is consequently a basic dimension of the
borderline patients ego weakness. Thus Zetzel and Kernberg seem
to be in agreement regarding the dominance of splitting mechanisms

and the importance of the rage or unneutralized aggression in the
borderline patient.
Similarly, the idealization of the therapist and the shift back
and forth between the idealization and the devaluation of the

therapist and between overvaluation and devaluation of the self are
explained on the basis of the splitting mechanisms and of the

mechanisms of projection and projective identificationat once
expelling evil aspects out of the self and putting them on objects,
leaving the self good and strong, or again taking them in again thus
making the self weak, helpless, and evil. This is in sharp contrast to
Kohuts approach as described above.
Zetzels approach to the borderline patient is therefore based

on an attempt to help the patient deal with the impaired ego. Due to
this impaired ego, In the therapeutic relationship, magical

expectations, impairment of the distinction between fantasy and
reality, episodes of anger, suspicion, and excessive fears of rejection
are to be anticipated over a relatively extended period. This is a
result of what Zetzel calls a developmental failure. The cure for this
failure is to gradually establish areas of relatively autonomous
functioning which are more or less free of the toxic effects of evil and
destructive introjects.
There is a certain atmosphere in this quotation about the
effects of evil and destructive introjects that is reminiscent of

medieval theology. There is a certain personification involved which
is semantically, metapsychologically, and scientifically undesirable.
This is the reason why attempts are being made to reconstruct and
clarify these theories of early development in order to remove the
moralistic and emotional overlay brought about by discussion of such
things as evil, destructive and malevolent introjectsalmost as if they
were an infestation by the devil.
CHAPTER 8
Developmental Failure in the Borderline Patient
The developmental failure in borderline patients, to

summarize Zetzel (1971, 1973), includes a failure in each of the
following developmental tasks: (1) the achievement of definitive self-

object differentiation; (2) the capacity to tolerate frustration, delay,
separation, and narcissistic injury; and (3) the internalization of a
positive ego identification which serves as a basis for a basic self-
esteem and a relatively substantial sense of autonomy.
With respect to the first task, we see the difficulty in the
borderline patient of distinguishing between fantasy and reality

especially under stress. According to Zetzel this also becomes a
problem in placing the patient on the analytic couch, but there is an

early report of a panel on the borderline patient (Rangell 1955) in
which Zetzel herself points out that in England the couch is employed
somewhat more freely and universally with borderline patients than

it is in the United States; no undue effects have been reported. Zetzel
explains, It is possible that the difficulties here of the use of the couch
stem to some degree from the analysts fears of the consequences

rather than from the patient, for the latter, if free association is
becoming too threatening, will usually set up his own controls. This
has been my experience in many instances also, and I have reported

elsewhere on this in detail (Chessick 1971b).
I hasten to add that the use of the couch in the psychotherapy
of borderline patients is not a procedure for a beginner. It should only
be
undertaken by the skilled, experienced and well-trained

psychotherapist who is aware of the difficulties involved. In some
cases it is then definitely very helpful; in others, it has no effect one
way or another. One of the clinical phenomena most impressive in
differentiating the borderline patient from the schizophrenic patient

is that schizophrenic patients usually do very poorly in unstructured
and couch situations; they really fall apart or become unintelligible.
Borderline patients simply usually do not.
With respect to the second developmental task, the intolerance
of aggressive impulses is most typically shown, and both the

aggressive impulses of the patient and of others are not tolerated.
Zetzel sees the lack of capacity to deal with frustration, delay, and loss
as due to the problem of splitting, which makes the patient
particularly susceptible to the toxic effects of destructive and hateful
impulses and impairs their capacity to master ambivalence.
The third developmental path is seen as most impaired and

Zetzel speaks of an impairment of the capacity for identification,
that is to say, the patient is seen as unable to effectively internalize

whatever strength is to be had from the therapist. Zetzel sees this
limitation and basic impairment in the capacity of the ego as setting a
limit on the effectiveness of therapeutic effort regardless of how
stable, realistic and consistent the doctor-patient relationship has
become. Because of these problems, Zetzel concludes that borderline
patients should be seen infrequently in order to avoid mobilizing the
intensive ambivalence and unneutralized aggressive aspects which
always threaten to disrupt the treatment. This is in contrast to

Kernbergs view of treatment, although both authors base their
conclusions on what seem to be essentially similar metapsychological
descriptions of the borderline patient.
It is not really a metapsychological statement to say that the
borderline patient has a limited and vulnerable capacity to
internalize a sufficiently stable ego identification and thus gain some
level of stable and genuinely autonomous functioning. This

statement does not explain why this phenomenon occursit is more
of a clinical impression.
However, Zetzels quotation is consistent with Giovacchinis
(1967a, 1975) view: Disruptive introjects do not lead to ego
differentiation. They interfere with the development of specific areas

of adaptation. Such lack of development or maldevelopment may
prevent the patient from obtaining gratification from persons who

may be willing to help him. The patient is not able to utilize or
assimilate experiences which another person that does not have the
same type of constricting introjects finds indispensable for his
emotional development.
There is little understanding and agreement on the exact
method by which the introject is formed or on precisely how it

functions in specific states of ego development to either enhance or
impair the adaptative functioning of the ego and the synthetic
functioning of the ego to produce the subsequent impaired
development of a healthy sense of self.
The task of psychotherapy with the borderline and psychotic

patient becomes infinitely complicated by the fact that the patient
neither has a firm grasp of his own sense of self nor is able, because of
the introjects, to respond to supportive, kindly or benevolent

measures as we logically would expect a starved and lonely person to
do. It is now theoretically clear why attempts at directly gratifying the
borderline patient have been repeatedly shown to fail. Directly trying
to mother the borderline patient causes serious chaos and often

produces a paradoxical response, leading to frustration on the part of
those who originally approached the patient with benevolence and
good will.
Kohut divides the class I have described as borderline patients
into those who form stable narcissistic transferences and are thus
treatable by the method of formal psychoanalysis on the one hand,
and all the rest on the other. The metapsychological differentiation

here rests on the issue of the cohesive self. This is characterized in
detail by Gedo and Goldberg (1973). It must be made clear that the
fundamental and irreconcilable difference between Zetzel and
Kernberg on the one hand and Kohut on the other is in the area of the
importance and vicissitudes of narcissism. Zetzel and Kernberg see
this narcissism as a pathological formation which is used by the
patient to hide and deal with intense unneutralized aggressive drives
and to compensate the patient for profound disappointments in
childhood. When it is removed by suitable interpretation the
primitive aggressive drives appear or are projected and are analyzed
by interpretation.
Kohut, on the other hand, emphasizes narcissism as a form of

normal development which has been arrested in certain types of
patients. If they are to be treatable by formal psychoanalysis, these
patients must have the capacity to form a stable narcissistic

transference in the situation. If they do not have this capacity, then in
the psychoanalytic situation they regress to a fragmentation of the
self and the only form of treatment that makes sense is for the
therapist to provide unification of the self by being a consistent
reliable object for structure formation to the patient.
In these situations interpretative approaches have little to do
with what happens, since if the patient is lacking a cohesive self-
system and cannot conceive of others in terms of whole objects, then

transference and interpretation make no sense. Gedo and Goldberg
write: It is sounder to conceptualize these events as consequences of
the therapists entry into a patients narcissistic world as a
transitional object; this intervention serves to bind and integrate the

fragmented personality through gradual mastery of narcissistic
injuries. This experience is usually not the reliving of any past
relationship, however, but a real experience in the present which may

have had no precedent. If the unification of the self is accomplished
in this way, then there may be additional improvement by way of

further maturation of various functions towards secondary
autonomy.
Gedo and Goldberg argue that such therapy, in which
unification and pacification are essential, is nonpsychoanalytic. The
treatment in such patients is utilized as a transitional object, and this

is the key to the therapeutic success. The reason these therapies are
called nonpsychoanalytic is that transference and interpretation do

not really have an effect, although the therapist may think so; what is
really helping the patient is the utilization of the therapeutic
atmosphere as a transitional object around which the unintegrated

nuclei of the self may cohere in order to allow the patient to find a
solid sense of self and therefore a better capacity to distinguish
between self and objects.
Even after this has occurred, the patient is still saddled with
infantile narcissistic positions involving the split-off grandiose self
and the search for the idealized parent imago described above. The
treatment is that of formal psychoanalysis, but if the patient cannot

form a stable narcissistic transference, then the therapist must be
satisfied with the use of pacification and unification techniques.
There is no doubt that pacification and unification techniques

have a very important role in the treatment of the borderline patient.
Winnicott calls this good-enough holding the therapist provides,
and we will go into this when we talk more specifically about

treatment. Here we have an important clinical differentiation based
on different theories of approach to the borderline patient. We will
take up later in detail the question, Should the therapist allow the
patient idealization of the therapist without interpretation for a long
period, or should such idealizations be interpreted somewhat
vigorously as attempts to hide tremendous rage and aggression?
It should be added here that even though formal
psychoanalysis is advised for patients who can form stable
narcissistic transferences, Kohut has found it necessary to introduce

a technical modification into this analytic technique. He advocates the
acceptance of the patients idealization of the analyst without
interpretation for a long period. In this sense, the analyst offers

himself as a new and real object for the purpose of permitting the
mastery of a developmental defect. The fundamental hypothesis of
Gedo and Goldberg is that effective therapies of the borderline patient

are based on the ability of the therapist to serve as a focus around
whom the clusters of unintegrated nuclei may coalesce into an
integrated, cohesive self. With Kohut, Gedo and Goldberg apparently
believe that there is no basic metapsychological difference between

the schizophrenic patient in remission and the borderline patient.
Both patients need the unification of a fragmented psyche through the

continued availability of the reliable objectthe presence of a real
person or even of a reliable setting: It is sufficient to establish an
uninterrupted relationship. From this point of view it makes little

difference what the therapist interprets to the patient, for what is
most important is not the verbal content of the therapeutic
transaction but the consistency, stability and reasonableness of the
relationship to the patient.
Again we see a profound and irreconcilable difference between

Kernbergs and Zetzels approach on the one hand and Kohuts, Gedos
and Goldbergs on the other, both in terms of technical therapeutic
principles and metapsychological description. Kohut places the focus
on the narcissistic sector of the personality and introduces the
technical modification of accepting the idealization of the patient
without interpretation for a long time. Kernberg does not accept the
idea of separate developmental lines of narcissism, and he aims to
reach the level of an Oedipal transference neurosis in the treatment.
The narcissistic structures revived in the analysis are interpreted by
Kernberg as defensive elaborations against primitive rage and
against more mature object relationships. He attempts to resolve the
pathological narcissistic structures through insight into the primitive

mechanism of the splitting of the object into good and bad. He
implements this aim by a direct confrontation of the patient with his
splitting mechanism and by a consistent interpretation of all
manifestations of his narcissism as defensive (Ornstein 1974a).
According to Kernberg, the patient needs to become aware of

his need to devalue and depreciate the analyst as an independent
object to protect himself from the reactivation of underlying oral-
sadistic rage and envy and the related fear of retaliation from the
analyst. Even in the use of the term envy, the difference between
Kernberg and Kohut appears, because envy already implies the
capacity to sense another person as a separate object. If a person is
only a self-object, you dont envy that person; it would be like envying
your right arm! So here again the theoretical differentiation is
apparent: self and object differentiation is thought of as occurring

significantly earlier in development in Kernbergs theory.
Kohut, on the other hand, focuses on what he conceives to be
the patients inability to perceive the analyst as a separate object, and
therefore empathic appreciation of this inability is required rather

than interpreting this phenomenon as a defense.
It seems to me that the closest approach to the resolution of

this problem so far has been suggested by Wangh (1974). He feels
that the discrepancy between these two views is largely apparent.
The intrusions of aggressive drive manifestations described by

Kernberg are the inevitable residues of the failure to develop of the
ideal sequence described by Kohut. According to Kernberg,
pathological narcissistic grandiosity carries with it the rage that

accompanied the frustration of normal narcissistic grandiosity.
Wangh writes, In other words it seems to me that in clinical practice
we inevitably meet, confront, and uncover both those phenomena
described by Kernberg and those set forth by Goldberg. The degree

and quality of the rage, often covered by stand-offish grandiosity, will
determine the degree of pathology in the individual patient.

Therefore, the seeming discrepancy may occur not only from a
particular approach taken by a particular therapist as a function of
the personality of the therapist, but also from the range of patients
that is, their degree of sicknesswhich each encounters in his
practice.
To put it another way, the psychotherapist of the borderline
personality disorders has to keep in mind three foci in understanding

the patient. The first of these is the traditional problem of the
formation of classical transference, in which those areas of ego
function most advanced in the patient will participate. In addition to
this, narcissistic transferences may form as described by Kohut; at
any rate, varying degrees of narcissism will always be an important
focus in the psychotherapy of the borderline patient, whether this
narcissism is a defense against both primitive formations and object
love or not. Finally, the focus on unneutralized aggression, splitting,
introjects and projective identifications is vital; the therapist will be

dealing with intense depreciation and hostility from these patients,
which always threatens the therapeutic alliance. The therapist will
have to develop the flexibility to deal with each of these foci in an

appropriate manner as they arise and in addition be able to provide
pacification and unification when fragmentation becomes a serious
threat.
The relative preponderance of narcissistic structures in the

presenting personality of the patient or of splitting mechanisms with
a ready tendency to contempt, hostility and depreciation, as well as

projective identifications and so on, determine the kind of patient that
is being described from a clinical descriptive point of view. Whether
or not the idealized parent imago and the grandiose self represent
way stations in the normal and separate developmental line of
narcissism or whether fixations in such formations are always
pathological and defensive cannot be settled at this time.
In view of the therapists having to deal with so much
unneutralized aggression, it does seem that it is wiser, at least in the

intensive psychotherapy of such patients, to accept the idealization
and to watch for and interpret the shifts from the idealized parent
imago to the grandiose self when there is disappointment in the
idealization, rather than to sharply confront and break up the
idealization by interpretation. The latter procedure, it seems to me,
makes it harder for the patient to internalize the idealized therapist,
which will be necessary to modify his attitude toward himself, and it
will increase the rage with which the patients already overburdened
ego has to contend.
For a long time those patients who present primarily with
unneutralized aggression and depreciation and envy of the therapist
need to be helped to deal with these phenomena through an

identification with the nonaxious, tolerant and consistent approach of
the therapist. Those patients who present predominantly with
narcissistic pathology (often including a tendency to fragmentation
and the formation of at least a hint of cold paranoid grandiosity and a

sense of persecution) need to be helped toward an insight into their
narcissistic pathology and toward an understanding of what the

narcissistic pathology defends them against and/or toward an insight
into how phase-inappropriate disappointments in childhood
generated a developmental fixation into the narcissistic pathology.

There is no reason why both of these phenomena cannot occur in a
patientthus a pathological grandiose self could be based on a
developmental fixation in the area of the grandiose self but could also
have a defensive meaning which would hypercathect and distort even

the normal grandiose self. I conclude this chapter by warning the
reader that these authors are using grandiose self in
metapsychologicallv quite different ways.
Part III
DEVELOPMENTAL PATHOLOGY
CHAPTER 9
Early Ego Development and Projective Identification
It would not be in the least surprising or perturbing to

discover that by this time the reader is somewhat confused with the
plethora of theories, all of which seem to be inconsistent and

sometimes downright conflictual, that have been introduced in an
attempt to understand early ego developmentwithout which, of
course, there can be no understanding of the borderline patient. So at
this point I am going to pause and try to make some kind of a
summary and side-by-side comparison of the principal theories that

have been put forward in a capsule form, so that the reader can
compare and contrast the various points of view.
Let us begin with the most objective observational kind of

theory. Mahler has of course done meticulous observational work on
the young infant and child. Using Mahlers timetable, from age zero to
two months occurs the normal autistic phase during which the
neonate is observed to be incapable of perceiving anything beyond his
own body. He cannot distinguish himself from his mother and seems
to live on a purely instinctive basis in a world composed solely of

inner stimuli.
During this phase the mother serves as an external executive
ego, replacing the childs initial incapacity to bind instinctual energies
and to delay discharge (Mahler 1952). She must prevent the neonate
from being overwhelmed and traumatized by internal stimuli and
help him achieve gradual transition from an exclusive cathexis of
processes within the body to an ever-increasing cathexis of sense

organs on the surface of the body and thereby to an increased sensory
awareness of the outer world. The great problem of this phase, then,
might be labeled the threat of overwhelming traumatization from

being flooded by unneutralized stimuli.
Observationally speaking, the second phase of life is labeled by
Mahler the symbiotic phase and lasts from around two or three
months of age to around six or eight months of age. During this phase,
which is marked by the infants beginning capacity to perceive at least
fleetingly that satisfaction is dependent on a source outside his body,
the mother is not yet perceived by the child as a specific whole

person. She can still be replaced by a substitute. The specific smiling
response at the peak of the symbiotic phase indicates that the infant is
responding to the symbiotic partner in a manner different from that
in which he responds to other human beings. He clings to a specific

symbiotic relationship with the mother. This specific symbiotic
relationship reaches its peak at about six months, and between six
and ten months of age occurs the beginning of what Mahler defines as
the separation-individuation phase.
The process of separation-individuation refers to a
psychological growing away from the undifferentiated symbiotic
relationship with the mother and a growing toward differentiation of
the self from the mother. Thus at the very time when the infants
specific attachment to his mother is growing stronger and stronger,
paradoxically his developing psyche and soma require him to begin to
detach himself from her in a series of separating and individuating
steps. This separation-individuation is completed, according to
Mahler, by about three years of age.
The separation-individuation phase is divided into subphases
by Mahler and her co-workers (Mahler and LaPerrier 1965, Mahler et

al. 1975). Although the chronology varies somewhat in her work, it is
important to be aware of these subphases:
1. Differentiation is characterized by increase in partial

locomotion and much scanning; he begins to express active pleasure
in the use of his entire body, shows interest in objects and in pursuit
of goals, and turns actively to the outside world for pleasure and
stimulation.
The differentiation subphase of the separation-individuation
phase lasts from about six to ten months of age, and during this
period there is maturational growth of locomotive function, active
pleasure in the use of the body and a turning actively to the outside
world for pleasure and stimulation, but these emerging functions are
still expressed by the child in close proximity to the mother, and we

begin to have the appearance of the well-known stranger anxiety, or
eight-month anxiety (Spitz, 1965), which appears around this time.
In this phenomenon, the percept of the strangers face is compared

with memory traces of the mothers face. It is found to be different
and rejected with disappointment: What he reacts to when
confronted with a stranger is that this is not his mother; his mother
has left him.
The continuous investigation of the mothers features in a
prolonged and sober visual and tactile exploration, and the comparing
and checking this with the features of others, seems to testify the
beginning of self and object differentiation.
2. The Practicing Period from about ten to eighteen months is
characterized by great strides in locomotion. At the peak of this
subphase, when the infant is one-and-a-half years old, the sense of

inflated omnipotence, the idealized state of the self, is at its height
and, in Piagets terms, sensorimotor intelligence just begins to be
replaced by representational intelligence. The first eighteen months

of life lead, in a sense, to upright locomotion and to this grandiosity or
self-inflation; the second eighteen months represent a corresponding
deflation.
Arieti (1974) points out that Piagets cognitive descriptions
are always treated by other theorists as if they were autonomous ego

functions, but in order to understand schizophrenic phenomena,
cognitive descriptions cannot be treated as if they were autonomous
ego functions. Their vicissitudes must be brought into our affective
theories of early development.
There is an additional benefit to mentioning Piaget at this

point, because if one follows his work (Piaget and Inhelder 1966;
Pulaski 1971) it is clear that the utmost caution must be used to
impute to the infant younger than two years old any capacity for
symbolic representation of good or bad object representations or

any kind of evocative images or representations for that matter!
Before the infant is one-and-a-half to two years old, according to

Piaget, cognitive development is characterized by coordination of
action in the absence of representation. During this so-called

sensorimotor period of Piaget, there is no symbolic activity and no
evocative memory, and adaptation is based on recognition in action of
familiar sensorimotor schema being experienced at the moment.

The burden of proof, therefore, falls heavily indeed upon those who
would postulate anything more than Piagets studies have indicated.
3. Rapproachemeni begins as the child learns to walk and lasts
from about sixteen to twenty-four months, the end of the second year
of life. The very pleasure of mastery is followed by separation anxiety
in the toddler as he becomes aware in the middle of his second year of
his true physical separateness. With this awareness, he begins to lose

his previous resistance to frustration and his relative obliviousness of
his mothers presence, and so he clings possessively to her.
4. The Fourth Subphase, essentially the third year of life, is

characterized by unfolding of complex cognitive functions; verbal
communication, fantasy, and reality testing.
When we move into the metapsychology of these

observational phases, it is important to understand that we have
made a sharp shift in position (Kohut 1971). As explained previously,
Mahlers position is defined as that of the observer who is equidistant
from the interacting parties and who occupies an imaginary point
outside of the experiencing individual. In metapsychology, our
position is that of the observer who occupies an imaginary point
inside the psychic organization of the individual with whose

introspection he empathically identifies.
Freuds famous phases of libidinal development roughly
correspond to Mahlers observations. Thus the stage of autoerotism

corresponds to Mahlers phase of normal autism from about birth to
two months. The phase of primary narcissism lasts during Mahlers
symbiotic phase from about two to six months, and the beginning of
rudimentary object love in Freuds sense appears at the beginning of

Mahlers separation-individuation phase.
Kohut makes serious modification of this. He essentially goes
along with the phases of autoerotism and primary narcissism of
Freud. He explains that around the beginning of the separation-
individuation phase, from about six to ten months, there is also a

transformation of primary narcissism along its own line of
development, first appearing at this point as the grandiose self and

the idealized parent imago already described.
By the time the separation-individuation phase of Mahler is
completed, that is at about the age of three, the formation of the

cohesive self, according to Kohut, is ready. Rudimentary object love is
possible, the transformations of narcissism have begun in such a way
that the grandiose self and idealized parent imago have started to be
substantially internalized, and a cohesive sense of self has been
formed. The individual becomes capable of beginning mature object
love, and the rudimentary primary narcissism becomes eventually
transformed into as such things as humor, wisdom, serenity and so
on.
Modells theory is somewhat more simple. His phase of no
selfobject differentiation roughly corresponds with Mahlers autistic
phaseabout zero to two months. A transitional object phase follows.

The objects during this phase are not perceived in accordance with
their true or realistic qualities, and he compares this with Winnicotts
(1951) discussion of the transitional object. Winnicott dates the phase
of the transitional object at four, six, eight or twelve months, that is to

say, the phase of the transitional object occurs around the beginning
of the separation-individuation phase and lasts through that first
differentiation subphase. Object love becomes more mature,

according to Modell, when separation and individuation are
completed, again in Mahlers terminology at about three years of age.
Kernberg (1966, 1972a, 1973) distinguishes four stages of
early ego development, briefly outlined here and discussed in detail

later. His first stage precedes the establishment of the primary
undifferentiated self-object constellation and corresponds roughly
with Mahlers phase of autism from zero to two months. In his second
stage, which corresponds roughly to Mahlers symbiotic phase from

about two to six months, there is an undifferentiated self-object image
(or representation) present. The third stage, which corresponds to
Mahlers differentiation subphase of separation-individuation, from
about six months to ten months, is when good and bad self-object
images are differentiated. At this point a stranger is greeted with
anxiety, and Kernberg interprets this as the first efforts on the part of
the infant to externalize the bad self-object image (compare the
explanation of Spitz given in Chapter 8). The fourth stage, according
to Kernberg, occurs between one and two years of age, around the
time that cognitive object permanence occurs. Affective object
permanency perhaps occurs a bit earlier. According to Kernberg,

affective object permanency occurs through the coalescence of good
and bad selfobject images into images that begin to correspond to the
real object out there.
Finally, we have the English school, for example, Balint, who
claims that rudimentary object love with primary cathexes to primary
objects is present shortly after birth and gradually extends into
mature object love. More theoretically, Fairbairn claims that libido is

simply a function of the ego and the ego is fundamentally object-
seeking. This is presented as a total and complete supplanting of

Freud, so one must choose between Freud and Fairbairn. Klein
postulates sophisticated object relations established shortly after
birth.
Modell and Kernberg essentially agree that the late symbiotic
stage or the early separation-individuation stage (Mahlers
differentiation subphase, from about six months to one year) is where

the damage is done in the borderline patient. Kohut places the onset
of narcissistic disorders at around the age of one year, lasting perhaps

to three years, at a time when the grandiose self and the idealized
parent imago are supposed to begin to be substantially internalized
and integrated into the ego and the superego. If this process does not
take place correctly, the developmental arrest leads to the formation
of the narcissistic personality disorder.
The appearance of classical eight-months anxiety, or stranger

anxiety, in the development of the infant marks the achievement of
cognitive self-object differentiation, a physiological as well as a
psychic achievement. At this point there begins an appropriate

coalescence and separation, according to Kernberg, of self-and object
representations and the development of rudimentary object love.
Kohut (1971), on the other hand, speaks of this as the era of

separation anxiety. The fact that cognitive self and object
differentiation becomes possible at around eight months of age,
according to Kohut, does not mean that a coherent sense of self has
occurred. Only nuclei of self-representations begin at this time.

Certain intermediate phases still have to be passed through.
Thus between around eight months of age and three years of

age, Kohut postulates an intermediate phase of powerful cathexis of
the grandiose self and the idealized parent imago. These psychic
formations are gradually internalized and integrated within the

psychic structure, and by the age of three the grandiosity becomes
confined to phallic narcissism; at the same time a cohesive sense of
self forms. With the resolution of the Oedipus complex at around six
to eight years of age, there is superego formation; moral anxiety
replaces castration anxiety. At this time the repression barrier is

established and consolidated, and after eight years of age anxiety
becomes confined to function as signal anxiety (Gedo and Goldberg
1973). From the age of eight years, further transformations of
narcissism occur, as well as a separate development of object love.
From the development of cognitive differentiation between the
self and object at eight months and until the attainment of a cohesive
sense of self at age three, the object is utilized as part of the childs
narcissistic world. Only after the establishment of a cohesive sense of
self can object love begin to develop.
Until the cohesive sense of self develops, disavowal or splitting
of the ego, as Freud calls it, is a major and important defensein
Kohuts terms, the splitting of the self. What happens in regression to

periods before the cohesive sense of self has developed is a
fragmentation of the self. Kohuts concept of fragmentation of the self

is crucial to understanding psychotherapeutic technique.
It follows from this that borderline patients are not treatable
by classical psychoanalytic methods. There is no cohesive sense of
self, there is no self-object differentiation, the self is fragmented, and

only pacification and unification (so that the ego nuclei can coalesce)
are possible.
On the other hand, when one postulates intrapsychic images

and representations and introjects as occurring during the phase
when the damage has been done that produces the borderline patient,
then the psychoanalytic method makes sense. For then the projection,
projective identification and reintrojection involving these
intrapsychic images can be interpreted to the patient and so worked
through and understood by him. Again, it is clear that there is an
absolutely irreconcilable difference involved here, between

theoretical preconceptions and corresponding treatment
recommendations.
I am ruling out the English school. Kleins use of various vague
concepts has already been discussed, and her discussion of the

paranoid and depressive position (postulating the death instinct as
the motivation for the anxiety behind these phenomena) seems to me
to be a considerable stretching of philosophical concepts and to
engage us in a semantic confusion. Her idea is that the infantile ego is

in danger of disintegration unless it can extrude destructive parts of
itself onto the maternal object, who is then seen as a persecutor; this
is known as the paranoid position. Later, to preserve the maternal

object itself, the object is split into good and bad portions, and then
we get the depressive position. It is difficult to accept the idea that
these complex intrapsychic processes occur shortly after birth!

Similarly, Fairbairns concepts require a total abandonment of all the
basic premises of psychoanalysis developed by Freud and a complete

new theoretical orientation; this makes them impossible to accept.
Modell (1975a) presents a heroic attempt to bridge this gap between
Fairbairn and Freudhis effort is the best in the literature so far.
We do have to look with greater care, however, at Kleins

concept of projective identification. This is used in many places in the
literature, and no two authors use it alike! Originally it was meant to
mean a forceful penetration in which the object was actually in

phantasy either injured or turned into an enemy. It was used also to
explain the emotional states that some patients may produce in the
therapist. This affect is claimed to exceed even countertransference
and has to do with the manifestation of the most primitive means by
which a baby can communicate emotions to its mother. If they are
disagreeable emotions, the baby can be experiencing relief by so
manifesting them. The motive is to evacuate or extrude the stress, and
the angry infant (patient) forcefully projects the hatefulness into the
mother(therapist).
Kernberg (1967, 1968, 1971, 1975a) repeatedly picks up the
concept of projective identification. The main purpose of projection in

the borderline patient, he points out, is to externalize the all-bad
aggressive self-and object images, and the main consequence of this
need is the development of dangerous retaliatory objects against
which the patient has to defend himself. This projection of aggression

is rather unsuccessful. As Kernberg explains, While these patients do
have sufficient development of ego boundaries to be able to

differentiate self and objects in most areas of their lives, the very
intensity of the projective needs, plus the general ego weakness
characterizing these patients, weakens ego boundaries in the

particular area of the projection of aggression. This leads these
patients to feel they can identify with the object onto whom
aggression has been projected and therefore increases the fear of
their own projected aggression. They have to control the object in

order to prevent it from attacking them, and they have to attack and
control the object before, as they fear, they themselves are attacked
and destroyed. At the bottom of this projective identification is the

lack of differentiation between the self and object, so that one
continues to experience the impulse as well as the fear of that impulse
while the projection is active and feels the urgent need to control the
external object.
It is a projection in which the projection has not thoroughly
worked, leaving the patients with a feeling that they must fear the
projected aggressor and at the same time that they have not really
been able to get rid of their own rage and aggression so they must
attack the projected aggressor also. They identify with this aggressor
that they project, and they must attack the aggressor just as they fear
to be attacked. They project out this aggression and they identify with
the aggression when it is projected out. This is a difficult concept but
is already modified from Klein, because at this point the projection

out of the aggression is not seen as a forceful penetration a la Klein, it
is seen as an attempt to deal with the enormous rage and aggression
of the borderline patient. Kernberg thinks this is the essential

difficulty and basic problem in treating such patients.
According to Kernberg, then, the projective identification is

understood as a consequence of the failure of the projection of so
much aggression in a person with a weak self-object differentiation.
On the one hand, he projects the aggression; on the other hand, he

identifies with the so-called aggressor that he himself has set up. This
leads to fear of attack and a need to attack and, of course, has obvious
crucial clinical consequences in terms of what happens between the
patient and the therapist in psychotherapy, both in transference and

countertransference. For example, In dealing with borderline
personality organization, dedicated therapists of all levels of
experience may live through phases of almost masochistic submission
to some of the patients aggression, disproportionate doubts in their
own capacity, and exaggerated fears of criticisms by third parties.
(Kernberg 1975a).
In borderline patients, according to Kernberg, the higher-level
ego structures are missing and early conflict-laden object relations
are activated prematurely in the transference. He tries to modify

Kleins theories because of her lack of precision and even notes the
confusion in her use of the term splitting (Kernberg 1972b). He does
accept a concept of introjection and projection somewhat related to

Klein, but he rejects the concept of introjection as having to do with
oral incorporation (it is never made entirely clear what Klein meant
by this anyway).
The ego states in the borderline patient represent an affect
linked to certain types of object representation images and certain

types of selfrepresentation images. Kernbergs concept of introjection
really has to do with a fixation of an interaction with the environment
forming an organized cluster of memory traces, leaving an image of
an object, an image of the self in interaction with that object and the
affective coloring of both the object image and the self-image.
Furthermore, it is postulated that these images occur quite
early. For instance, the well-known reciprocal smiling response at
about three months of age may mark the first beginning of an

organization of the psyche. It is important to realize, Kernberg
explains, that such formations and images can take place quite early.
He asserts that splitting as an active mechanism comes into operation
around the third month of life and reaches its maximum several
months later, only gradually disappearing in the latter part of the first
year of life. The later developments of the ego presuppose an

important overcoming of the splitting processes. Kernberg disagrees
with Klein and Fairbairn in their assumption that an ego exists from
birth, however.
In the borderline patient the splitting has never been resolved
and the weak ego falls back easily on the splitting, creating a vicious
circle in which ego weakness and splitting reinforce each other. This
leads to the observed clinical phenomena in the psychotherapy of the

borderline patient. Kernberg here and there also suggests a
constitutional factor of perhaps extreme aggression or a
constitutionally determined lack of anxiety tolerance which may

interfere with the synthesis of important introjects of opposite
balances, as I will explain later. This is never made quite specific in his
theory and remains vague as the postulation of a constitutional X
factor.
Wolberg (1973) stresses Kleins concept of projective

identification in the psychodynamics of borderline patients even
more than Kernberg. In general she considers the concept of ego
defect in the borderline patient as giving way to a new concept of

defense, in which the individual is shielded from perceiving the true
nature of the reality situation in the parental home. A
sadomasochistic role is assigned to a child by his parents. The child
becomes enmeshed as a transferential object in service of the
parents defenses. The ensuing neurotic and psychotic processes are
defenses against a harsh reality with which the individual tries to
cope, and the crucial mechanism is identification with the parents. In
projective identification there is passive and masochistic behavior
with sadism as its goal; the insistence on a sadomasochistic position is
a defense against intense oral aggression. The patients deny their
autonomy and, out of rage, force others to do for them what they
should do for themselves.
According to Wolberg, identification in this situation is not a
form of development or ego growth but a substitute for object

relations, and in psychotherapy this projective identification has to be
unravelled and broken up. For example, the therapist may be
regarded as a sadistic mother-image, while the patient experiences
himself as the frightened, attacked child; then, even moments later

the roles may be reversed. The goal of the therapy is to show the
patient that in his projections he sees in others what is also in himself.

This serves to externalize the introject and makes it possible to
discuss the unconscious motivations, the fantasies and the defenses of
the other. These goals are accomplished essentially by confrontation

at appropriate times. Thus, for example, a woman patient quoted by
Wolberg, denies the sadistic side of the controlling mothering
mechanisms and says that she is really good. She would never have
acted that way with her mother if her father had not made her do so.
She denies that she acts this way today with her husband, although in
previous sessions she has alluded to this.
What Wolberg calls projective interpretation is the first step in
outlining an interlocking defensive pattern between two people; as in
Kernbergs conception of narcissism, the purpose of the projective
identification is to defend the patient against both his own powerful
unneutralized aggression and his fear of destructive retaliation from
others as a response to this aggression and as a projection of this
aggression onto others.
All seem to agree that treatment involves modifying cold,

unloving and archaic ego and superego introjects with new warm,
loving and reasonable introjects. The patient learns an adequate way
of life through identification with the analyst, i.e. through analytic

introjects; this identification must be fostered if the patient is to gain
an understanding of reality. One must be very careful in emphasizing
this technique to avoid the use of confrontation as a disguised form of
countersadism on the part of the exasperated therapist, for everyone

agrees that narcissistic and borderline patients produce tremendous
countertransference reactions in the therapist, which vary all the way

from sarcastic putting down of the patient to actually acting out in a
massive retaliatory and destructive way toward the patient.
Kernberg (1975a) states that in addition there is quantitative

predominance of negative introjections stemming from both a
constitutionally determined intensity of aggressive drive derivatives
and from severe early frustrations. As a consequence, the child never

does give up the splitting and in turn creates a serious problem in the
development of the autonomous ego functions. Thus we have an

emphasis on bringing forth and interpreting the negative transference
in terms of the projection and on the breaking loose of severe oral
aggression, which is kept under control by a careful structuring of the
therapeutic situation through the use of so-called parameters.
A frank dealing with the manifest and latent negative
transference is absolutely necessary, and trying to avoid this under

the guise of building a therapeutic alliance leads only to a vicious
circle: Projection and reintrojection of sadistic self and object-images
in the transference. This is Kernbergs objection to Kohuts allowing

the idealization of the therapist to develop without interpretation. He
argues that this avoids the mobilization of the latent aggression,
which can only be worked through by its becoming manifest in

projections and direct hostility toward the therapist. Unless this basic
aggression is allowed to surface and is worked through in the
psychotherapy, no fundamental structural change can take place in
the ego of the patient, since all the energies are bound up in dealing
with this archaic, magical and frightening aggression.
Chapter 10
Internalized Object Relations
For Kernberg, projection, introjection, projective identification

and pathological narcissistic formations are all defense used by the
patient to deal with aggression. The narcissistic personality structure

then becomes a form of borderline personality organization in which
the primary operation for maintaining the splitting is the formation of
a highly pathological grandiose self. The integration of this
pathological grandiose self . . . explains the paradox of relative good
ego functioning and surface adaptation in the presence of a

predominance of splitting mechanisms, a related constellation of
primitive defenses, and the lack of integration of object
representations. (Kernberg 1974a).
Again, the disagreement is about the origin of this grandiose
self whether it reflects the fixation of an archaic primitive self in the
course of normal development or a pathological structure clearly

different from normal infantile narcissism. Those patients who
function on an overt borderline level characteristically show
repetitive chronic activation of intensive rage reactions replete with

demanding and depreciatory attacks on the therapist; the
pathological grandiose self enables the patient to avoid this, but when
it is broken up through interpretation the rage appears again.
The decision for Kernberg as to whether the borderline patient
is amenable to uncovering intensive psychotherapy rests, therefore,
not on whether or not stable narcissistic transferences form, but on

whether the nonspecific manifestations of ego weakness are so great
that they rupture the treatment. Thus, in the presence of a severe lack
of anxiety tolerance, generalized lack of impulse control, absence of
sublimatory channeling and strongly predominant primary-process
thinking, as well as a tendency toward delusion formation, we have

the danger of transference psychoses and destructive acting out
which can only be treated by a supportive and authoritative
psychotherapeutic approach, at least until structure has been

provided for the patient that holds these reactions within workable
and socially acceptable limits. After this has been accomplished, it
may then be possible to go forward in an intensive uncovering

psychotherapy. The outbreak of dangerous, aggressive paranoid rage
as well as intense depression and guilteven with the possibility of

suiciderepresents the important prerequisite of the working-
through process but can only be allowed if the therapist is confident

that the patient is able to keep these manifestations within the safe
boundaries of the psychotherapy.
Notice again the main thesis that the structures determined by

the internalized object relations constitute a crucial determinant of
ego integration, and an abnormal development of internalized object

relations determines varying types of psychopathology. Kernberg
(1972a) outlines four stages of development of internalized object
relations which, if one can accept the early existence of such

intrapsychic structures, is the least objectionable and confusing of all
the theories. These stages are, as already mentioned briefly in
Chapter 9, as follows:
Stage One, in which the primary undifferentiated self-object
constellation is built up under the influence of pleasurable gratifying
experiences of the infant in interactions with his mother, occurs

somewhere in the second to third month of life.
Stage Two consists of the establishment and consolidation of
an undifferentiated self-object image or representation of a
rewarding type under the organizing influence of the gratifying
experiences of the child-mother unit. Thus, A primary intrapsychic

structure is built up, with early traces fixating the primitive
coenesthetic constellation and its gratifying all good affective
quality; this constitutes the primary, undifferentiated, self-object

representation. Simultaneously, a separate intrapsychic structure
representing an undifferentiated all bad self-object representation
is built up under the influence of painful and frustrating

psychophysiological states. At this stage there is no separation
between self and nonself.
Stage Three is reached when self and object have been

differentiated under the influence of perceptual and cognitive growth.
Ego boundaries stabilize and self-images become separate from
object images, but there is not yet an integrated concept of the self, for
good and bad images are separate. Thus, object constancy is not yet
possible.
Stage Four, the final stage of integration of good and bad self-
images occurs in the second year of life with a coalescence into an
integrated self-concept. Affects become integrated, toned down and

differentiated. At the same time a coalescence of good and bad object
images takes place, fostering better discrimination among object

images stemming from interpersonal relationships and a more
realistic representation of significant others. Thus, An integrated
self-concept surrounded as it were, by an integrated conception of

others, with ongoing modifications of self concept and concept of
others in the process of interpersonal relationships, constitutes ego
identity in the broadest sense.
The transition from Stage Three to Stage Four is where the
disaster has occurred in the borderline patient. Differentiation of self-
images from object images has occurred to a degree sufficient to
permit the establishment of integrated ego boundaries and

concomitant differentiation between self and others. However, the
coalescence in integration of good and bad self-images and object

images fails because of the pathological predominance of primitive
aggression. The intensity of aggressively determined self-and object
images and of defensively idealized all good self-and object images

makes integration impossible, for bringing together extremely loving
and extremely hateful images of the self and significant others
produces an unbearable anxiety and guilt. The result is active

defensive separation of such contradictory images, and splitting
becomes a major defensive operation.
The integration of loving and hateful feelings in the context of
internalized relationships with others is a major precondition for
neutralization of instinctual energy, according to this theory. Lack of
such neutralization deprives the ego of an important source of
sublimatory potential and of the conflict-free emotional sphere. It
brings about a persistence of primitivization of emotions with
secondary lack of impulse control. Thus premature sexualization and
oedipalization of relationships with parental figures leads only to

aggressive contamination of the sexual life of the patient.
Similarly, correct superego development cannot occur, leading
to an overdependency on external sources of reassurance, praise and

punishment. The channeling of aggression into the psychic apparatus
from which the ego and the self develop serves the biologically
protective function of avoiding external discharge onto the mothering
figure upon whom the infant is so dependent. This inward direction of

aggression is normally elaborated into stable internalized object
relations that successfully neutralize the aggression, but this

mechanism fails in patients with borderline personality organization.
It is not clear whether it fails because of a constitutional weakness or
defect or because of the profound phase-inappropriate

disappointment that takes place at the time, generating larger
quantities of rage and aggression than the psychic apparatus is able to
handle.
The remobilization and discharge of this aggression in the
therapeutic situation lightens the burden on the psychic apparatus

and permits the eventual fusion of good and bad self-and object
images, providing the remobilization does not lead to psychotic or
destructive behavior that ruptures the treatment entirely.
Kernberg, as stated, feels that the grandiose self described by
Kohut is always pathological and that even the idealizing transference
of Kohut is simply the projection of the grandiose self onto the

therapist. This pathological configuration, according to Kernberg,
defends the patient against profound oral rage and envy, paranoid
fears (due to the projection of sadistic trends onto the therapist, who
becomes the hated and sadistically perceived mother image),
loneliness and hunger for love and guilt over aggression. Thus one of
the big clinical questions is whether the grandiose self as it appears in

the patient and develops in the mirror transferences should be
allowed to remain until it is gradually broken up and, through
transmuting internalizations, be brought into integration in the
personality or should be directly interpreted as a defense.
In pressing his point of view about the grandiose self,

Kernberg (1974b) differentiates infantile narcissism from adult
pathological narcissism in at least five ways: (1) the childs
narcissistic demands are more realistic; (2) in the child these
demands coexist with object love when they are not frustrated; (3)
the childs demands relate to real needs; (4) there is a certain warm
quality about the childs self-centeredness; and (5) the exclusivity and
totality of wishes for admiration, wealth, power and so on are far

greater in adult pathological narcissism. The child wants more in the
way of loving and sharing.
The combination of high inborn intensity of aggressive drives
and a mother who is cold, narcissistic, overprotective and includes the
child in her narcissistic world gives him a sense of specialness around
which the grandiose self crystallizes and which leads to a pathological
grandiose self-formation in adult patients.
Thus the intrapsychic world of the borderline patient with
these kind of problems (Kernberg 1974b) consists solely of a
pathological grandiose self, devaluated shadow images of the self and

others, potential persecutors which are nonintegrated, sadistic
superego forerunners and primitive distorted object images onto
whom the intense oral sadism has been projected. Laboring with this
nightmare in the intrapsychic world, a pathological grandiose self

provides better social adaptation than in the usual borderline patient
and compensates the patient for inner suffering, but the price is a
grandiose isolation and a loss of true human contact. In addition to

that, adult superego formation is prevented, so a sadistic superego
represents a constant danger to the ego, even to the point of self-
destruction.
Thus, Kernberg sees a lack of interpreting of the mirror

transference and idealizing transference as supportive tolerance of
the narcissistic constellation, while Kohut implies that to insist that
the idealizing transference hides profound hostility and exists solely
for that purpose represents a countertransference in the therapist!
Finally, the presence of all good and all bad object energies
which cannot be integrated interferes seriously with superego

integration as follows (Kernberg 1967): Primitive forerunners of the
superego of a sadistic kind, representing internalized bad object
images related to pregenital conflicts, are too overriding to be
tolerated, and are reprojected in the form of external bad objects. We
see the same problem underneath the pathological narcissistic

configurations, for at the bottom of these pathological structures is a
hungry, enraged, empty self, full of impotent anger at being
frustrated, and fearful of a world which seems as hateful and

revengeful as the patient himself (Kernberg 1975a).
Chapter 11
Unresolved Metapsychological Problems: My Views
We turn now to a resolution of the differences between two

eminent groups of psychoanalysts on very important issues, from the
point of view of the psychotherapist who practices predominantly

intensive psychotherapy. We need answers to the following
questions:
1. Is idealization of the therapist a defense against rage and a

projection of the pathological grandiose self, or is it a
search for the idealized parent imago due to a
developmental arrest?
2. Does narcissism undergo a separate developmental pathway

and do narcissistic disorders thus represent an arrest
of development, or does the presence of the grandiose
self simply represent a pathological structure
developed to defend the patient against profound rage
and envy and so on?
3. Are the differences between borderline patients and

narcissistic personality disorders fundamental or not?
That is to say, is the narcissistic personality disorder
separate from the borderline patient and treatable by
formal psychoanalysis, or is the borderline patient
simply a psychotic who is not manifesting overt
symptomatology and who is not treatable by formal
psychoanalysis or modified psychoanalytic
psychotherapy?
4. Should the mirror and idealizing transferences (Kohut) be
interpreted as defenses against negative transference
or should they be accepted as manifestations of
developmental arrest which
5.will be spontaneously broken up through the process of the

psychotherapy and reintegrated through transmuting
internalizations?
6. Are Kernbergs outline of the phases of early ego integration

and object relations and his postulation of the basic
pathology in the borderline as that of splitting and the
presence of nonintegrated good and bad self-and
object images correct?
7. Is the grandiose self shown by the adult patient in

psychotherapy a developmental arrest of the normal
childs grandiose self or is it fundamentally different
than the grandiosity of the small child?
8. Is there a fundamental difference between the borderline

patient and the psychotic patient with important
ramifications for the psychotherapy of each?
9. As Ornstein (1974) asks, are these experts talking about the

same level of observation or are they not even talking
about the same patient population?
It seems to me that this last question may contain the secret of

the discrepancy. Perhaps it is clarified when we look at the way that
the two authors deal with the problem of aggression. Ornstein writes,
For Kernberg, the raw id-derivative of aggression in the form of oral

rage and envy is inevitably intertwined with the earliest projectively
and introjectively internalized object relations. . . . For Kohut, the id-
derivative of aggression becomes activated or psychologically
elaborated as it arises experientially from the matrix of archaic
narcissism.
From the review of my own clinical material I have come to
certain conclusions, and I will state these conclusions in advance for

clarity. They are:
1. I think that two different patient populations are being

discussed by Kohut and Kernberg and that this is
causing a lot of the confusion. On the whole, the kind
of patient that comes into a psychoanalytic institute
asking for formal psychoanalysis or at least gets
through the institutes sophisticated intake
procedures and is referred to a certified
psychoanalyst is different from the run-of-the-mill
borderline patient who comes in seeking
psychotherapy from therapists who are struggling to
make a living out in the field.
2. There is no reason to believe that the kinds of transference

formed have to be restricted either to the purely
narcissistic stable transferences described by Kohut
or to regressive fragmentation. It is possible that other
kinds of workable transference can form, for example,
the transitional-object type of transference, as
described by Modell, and this can be more
characteristic of the run-of-the-mill borderline patient
and yet stable enough to lend itself to
psychoanalytically informed intensive psychotherapy,
if not to formal psychoanalysis.
3. In general the theoretical conceptions of Kohut seem to be the

more acceptable and believable. They dont postulate
as much sophisticated, intellectual and cognitive
functioning for the small baby, and they allow the
formation of the narcissistic disorders to take place a
little later. Although Kernberg has modified some of
the Kleinian concepts, his theoretical formulations still
assume a considerable sophisticated capacity on the
part of the six-or seven-month-old infant to form and
retain self-and object representations.
4. There is no doubt that the splitting Kernberg describes seems

to be present in the adult clinical material. The
question is whether what is described as due to
splittingthat is to say, the projecting out of
malevolent introjectsis not really a telescoping that
the patient undertakes for the purpose of
communicating primitive affect. In this way, later
feelings and later self and object representations,
which form after the stage of the cohesive self, are
invested with primitive and early affects, especially
aggressions, for the purpose of communicating and
externalizing the unbearably intense current affects of
rage and so on.
5. On the other hand, Kohuts theory has certain unsatisfactory

aspects to it, because it is only half a theory, as he
himself admits. In Kohuts monograph (1971) he
deliberately discusses the libidinal aspects of
narcissism. Only in a paper (1972) does he discuss
narcissistic rage. The enormous rage in these patients
is not sufficiently explained by the theory of
narcissistic injury. There is something not quite
worked out in Kohuts formulations, and we look
forward to his later and more complete formulations
of aggressive aspects of narcissism.
The answer to all of these questions can only be found by
reference to clinical material. For some patients the descriptions of
Kernberg seem very appropriate, whereas for others Kohuts
approach seems much more to the point. The dividing line seems to
have fundamental significance for psychotherapy in terms of the
clinical appearance of raw aggression in the form of oral rage and

envy with an apparent splitting into good and bad self-and object
representations.
It is usually possible to identify patients who are using

narcissistic defenses against primitive disintegration as a
consequence of oral aggression, because these forms of aggression
intrude themselves in a variety of subtle and sometimes not-so-subtle
demands in spite of the narcissistic formation. Usually this type of

patient is more accurately labeled the borderline personality
organization and appears more fundamentally to fit the
psychodynamic descriptions of Kernberg.
You will notice that in describing my cases I am using
Kernbergs psychodynamics wherever it is appropriate, but please

remember that I think a telescoping goes on, so that when these
patients seem to show projection of good and bad self-and object

images, it is not necessary to conclude that that comes from what
went on in the infants psyche at age six months. Rather, what the
therapist is experiencing is the communication of extremely primitive
affect, which is presented with ideation that is borrowed from a later

phase of development at a time when there is adequate cognitive
capacity, including the capacity to form, retain and re-present self-and
object images.
The primary problem in borderline patients is really founded
upon the development of enormous undifferentiated primitive rage that
goes way back to the patients earliest days. Such overwhelming

negative affect disrupts the development and smooth functioning of the
psychic apparatus. Its origin can be traced to a devastating enemalike

intrusiveness and massive inconsistency primarily based on lack of
empathy on the part of the mother. It is then later attached to and
appears clinically in phantasies and projections of destructive archaic

bad unintegrated self and object images, among others.
There also seems to be a type of patient described by Kohut
who functions at a better and more integrated level and is not

struggling so pathetically and constantly with these intensive early
unintegrated affects. As Kohut has pointed out, attempts to interfere
with the development of mirror or idealizing transferences in such
patients, for example by insisting to the patient that such

transferences hide hostility, do produce hostility, but a hostility not
attached to structures that were hidden by the narcissistic pathology;

it represents rather a withdrawal into narcissistic rage as a
consequence of lack of empathy from the therapist. This kind of
interpretation breaks up the treatment primarily because it is not

correct!
It is very important to make as clear as possible a distinction

between the two kinds of patients, because in one kind, the rage
produced when the grandiosity is interpreted is appropriate in the

therapy and facilitates a release of material previously repressed,
whereas in the other kind, the rage produced signals a failure of
empathy on the part of the therapist; the consequent narcissistic rage
is a repetition of what happened between the patient and his mother.
I believe the difference in patient populations is the solution to the
disagreement between the authors under discussion, and it is a very
important clinical distinction. I also fully realize that this solution
leaves many metapsychological questions unresolved, and I cannot
attempt to resolve these problems here. The reader is referred to the

increasingly sophisticated metapsychological discussions in the
literature that attempt to distinguish the borderline patient from the
narcissistic personality disorder (Modell 1975a, 1975b), and I will

offer only some clinically pertinent comments.
An increasing focus on the precise failures in mothering that
produce these two kinds of patients is taking place in the literature.
The concept of the intrusiveness of the mother was discussed by

Heimann (1966) a number of years ago. She suggested that the bad
internal objects do not arise as a result of active introjection by the

infant, but as the result of passively endured intrusions of an unloving
mother, beginning during the undifferentiated stage when the infant
is maximally helpless. Mahler et al. (1975), in summarizing her

findings on the subject of middle-range pathology, emphasize how,
in the rapprochement phase of separation-individuation especially,
the more intrusive and unpredictable the mother is, the less the
modulating and negotiating functions of the ego gain ascendency.

Thus, predictable emotional involvement, consistency and minimal
intrusiveness are the primary ingredients of mothering necessary to
avoid fixation in a rapprochement crisis, using Mahlers terminology.

When such fixation takes place, coercive behavior, such as temper
tantrums to force mother to function as an omnipotent extension, and
desperate clinging, with a splitting of mother representations in order

to protect mother against aggressive drives, must take place, leading
to the eventual clinical picture of borderline pathology.
The patient is highly vulnerable to separation anxiety because

of the precocious hatching necessitated by the uncomfortable
symbiotic stage. Aggression tends to be aimed at the self in order to
preserve the mother, and a poor internalized mother representation
is available for inner sustenance under stress. Mahler emphasizes the

total body experiences (Spitzs concept of coenesthetic global
experiences) necessary for a successful symbiosis and the importance
of the internal mother, defined as the inner image or intrapsychic

representation of the mother, who in the course of the third year
becomes available as a soothing mechanism in the mothers physical
absence. In borderline pathology, in place of the internal mother there

exists a set of bad experiences intruded into the helpless infantile
egoa destructive substructure perhaps formed by cumulative
trauma (Kahn, 1974).
Masterson and Rinsley (1975) point out how Kernbergs and
Mahlers timing of the occurrence of the fixation underlying

borderline personality development differ significantly, the former
citing the period of 4 to 12 months and the latter the period of the
rapprochement subphase, coinciding with 16 to 25 months
postnatally. They agree with me that, the preponderance of

evidence would appear to be more favorable to Mahlers timing, but
they emphasize more the mothers withdrawal of her libidinal
availability as the child makes efforts towards separation-
individuation during the rapprochement subphase. The borderline
mother, according to these authors, has the defensive need to cling to
her infant, and therefore the childs separation-individuation
represents a major threat. I fear the tendency to oversimplify in this
viewpoint, leading to another devilish personification like the
schizophrenogenic mother.
Modell (1975a, 1975b) builds on the concepts of intrusion and
inconsistency in the most careful effort to date to distinguish, in this
middle-range pathology, between the borderline patient and the

narcissistic personality disorder. He sees all these disorders as
representing the psychopathology of object relations, related
primarily to an actual failure of the human environment. Thus,
bringing the above authors together, he writes, The environmental

failure may be massive and obvious, such as a failure of a constant
and reliable maternal object in the first and second years of life, or the
failure may take more subtle forms such as a failure of the mother to
accept the growing autonomy and individuality of the young child,
thus interfering with its sense of identity and separateness.
In Modells view, the more massive failures in the preoedipal
period leave the patient with an intense persistent object hunger and
lead to the clinical picture of the borderline patient. The subtle

failures allow the patient to internalize something, but due to a
premature disillusionment with the mother, a precocious and fragile

sense of self develops. This leads to the narcissistic character with the
false self, as described by Winnicott. He is led theoretically to a
revision of psychoanalytic theory and to a postulation of the
instinctual nature of object relations, which brings us once more to
the highly controversial metapsychological disputes reviewed
recently, for example, by Friedman (1975).
For the clinician, what is important to keep in mind are the
issues revolving around the terms defect and developmental arrest! I
believe those cases marked by massive failure in the maternal

environment show borderline pathology and are grounded on an
intrapsychic defect, primarily the lack of strong positive
identifications to neutralize and modulate aggression. Narcissism and

introject formation in such patients are related to the patients effort
to set up his own substitute structures in order to deal with
aggression and other drives and achieve some kind of adaptation to
life. The narcissistic personality disorder, on the other hand, has

achieved some internalized psychic structures, although these are
primitive, and is responding to a more subtle form of failure of the

maternal environment. Disillusion with the mother in a precocious or
phase-inappropriate manner is the central factor leading to
substantial developmental arrest in the area of narcissism.
This is a clinically vital distinction, because clearly the
therapeutic strategy for patients suffering from an intrapsychic defect
will be substantially different than for patients with a developmental

arrest; furthermore, these are theoretical poles and our patients
actually present with a combination showing the preponderance of

one or the other. The secret of successful treatment depends on
meticulous evaluation of the patient so as not to confuse defensive or
substitute structures with pristine or archaic psychic structures as
they manifest themselves in the patients personality and behavior.
Chapter 12
Clinical Material
Let me turn directly to a series of brief vignettes from my

practice to illustrate what I am talking about. Needless to say, this
discussion is of fundamental importance in approaching the intensive

psychotherapy of the borderline patient.
Patient #1:
Very soon after being placed on the couch he reported the

fantasy that I was going to cut his throat and kill him. At first I
became very alarmed by this fantasy, arising as early as it did in the
therapy, and sat him up for a year or so of treatment. As I got to know

this man I realized that he was actually quite intact and able to
function; there was no history of psychotic breakdown or behavior in

his past life, nor in any way could there have been attributed to him a
narcissistic personality disorder. He showed none of the
characteristic features of such patients, but presented rather with a

depression secondary to homosexual preoccupation with a friend
who had moved away. This was worked through in long-term
intensive psychotherapy, and gradually the rage deep in this patient

began to show itself again in the father transference and later in the
mother transference.
His early childhood was an absolute nightmare and the
psychodynamics of his homosexual interests and his depression were
clear. At the point where the early enormous oral aggression and envy
threatened to emerge in a more primitive form (as they had instantly
appeared in a delusion early in the therapy), the patient developed a

series of narcissistic defensive configurations related to his work
capacities and his sense of entitlement from the world. At this point
the clinical material appeared to be that of a narcissistic personality

disorder. In this patient the interpretation of the narcissistic
configurations as a defense against the frightening rage was
appropriate and helpful, and the patient was able to go forward

through a series of projections onto myself and his wife (during
psychotherapy his homosexuality had profoundly diminished and
was replaced by heterosexuality and marriage) so that his rage and

fear of his sadistic intrusive aggressive mother could be worked
through.
Patient #2:
A woman entered psychotherapy after her husband turned

from her to a perversion, leading to a subsequent divorce. The
presenting situation was one of a profound narcissistic injury and
deep overwhelming narcissistic rage as a response to her husbands

choice of a perversion over her. Here the picture described by Kohut
seemed extremely accurate. Even the patients acting out by sexual
promiscuity fit Kohuts (1971) formulation that the acting out in the
narcissistic personality disorder represents a process similar to
symptom formation in the neuroses. He writes, The acting out of
narcissistic personalities is a symptom which is formed in
consequence of a partial breakthrough of repressed aspects of the

grandiose self. Thus, although usually maladaptive and often
disruptive, it may nevertheless be regarded as an achievement of the
ego which amalgamates the grandiose fantasies and exhibitionistic

urges to suitable preconscious contents and rationalizes them,
analogous to the process of symptom formation in the transference
neuroses. This patient acted out sexually by conquering a whole

variety of men and at work by taking on an enormous burden of
responsibilities, behaving as if she were the president of the company.
The psychotherapy was marked by the characteristic
countertransference problems (described by Kohut) in dealing with a

mirror transference, which problems developed fairly rapidly:
boredom, difficulty in emotional involvement with the patient and
precarious maintenance of attention, leading to overt anger at the

patient, a tendency to exhortations, forced interpretations, and
tension and impatience on the part of the therapist. The patients
verbal and nonverbal behavior were not object-directed and
contained the demand for total enslavement of the therapist as a

prestructural object in order to help maintain her narcissistic
equilibrium.
Following Kohut, and with his insights, it was possible through
allowing the mirror transference to developin this case an archaic
merger transferencefor the patient to restore her narcissistic

equilibrium, relinquish the acting out, resume her social life and
remarry and return to her previous level of functioning. This satisfied
the patient, and no attempt at an extensive psychotherapy of her
narcissistic disorder was made, since the patient was pleased with
her former level of functioning.
Patient #3:
She presented as a typical borderline patient with no evidence
of a narcissistic personality disorder or symptoms centering on the
problems of narcissism. Kernbergs psychodynamics seemed very

appropriate, and the patient needed a lot of help in dealing with split
good and bad self-and object images, which were constantly projected
and reintrojected so that the patients life became an alternating cycle

of periods of relative peace and periods of rage and paranoid fears. At
no time were narcissistic defenses predominant in this case, and the
psychotherapy centered around providing an accessory structure to
the patient to help her deal with those periods when the pressures
from the unintegrated bad self-and object images disrupted her
functioning level. It was not possible for many reasons for this patient
to be treated at more than a supportive level, but this form of
treatment enabled her to function successfully and stop her
alcoholism. The essence of the technique was to bolster the egos
defenses and to support the less pathological defenses so that the

patient could deal with her aggressive and destructive impulses and
continue to function.
Patient #4:
This man presented as a schizoid personality who gradually
revealed a variety of narcissistic and exhibitionistic fantasies. He
came into psychotherapy because his marriage was failing and he was
beginning to realize that he could not ever achieve anything remotely
close to his primitive fantasies of total grandeur, admiration and
exhibitionism as a world-famous musical performer and a chess
genius. Nothing more was possible with this patient than to allow a
therapeutic alliance to develop and to become the patients only

friend and confidant, seeing him infrequently and sharing his
concerns as well as pointing out reality and providing some structure
to his life. This enabled him to function at his schizoid level but
removed his serious depression, which had dangerous suicidal
components to it.
To assume that under his profound narcissistic fantasies were
enormous fragmented bad and good self-and object images would be

simply a speculation; he never gave any evidence that this was a
problem. What he constantly presented was a deep longing for love;

when someone paid attention to him he was happy, and when he was
ignored by everybody he became depressed. Here I felt it was best not
to be a bull in a china shop, as Kohut describes it, and simply to

promote the patients functioning at his schizoid level.
Patient #5:
Over a period of many years she had had two or three acute
psychotic episodes with delusion formation of a paranoid nature and
innumerable hypochondriacal preoccupations. The tendency toward

fragmentation was continually present under stress. In spite of
therapeutic efforts the patient slowly drifted to a lower and lower
level of personality functioning. The great problem was the tendency
to fragmentation of the self under stress which displayed itself by the
appearance of the hypochondriasis, by delusional reconstitution of
the grandiose self in a cold, paranoid grandiosity, and by profound
narcissistic rage. There was a persistent theme of delusional
reconstitution of the omnipotent object in terms of a powerful

persecuting system against her. Even antipsychotic medication
seemed to have no effect on these fragmentations, nor did anything
else; it was as if the basic glue holding her personality and her self
together was defective. In this case the regression was to a stage prior
to that of the cohesive self. The narcissistic configurations appeared
to be a desperate attempt to protect herself against fragmentation, an

attempt analogous to her innumerable visits to innumerable doctors
for her hypochondriacal complaints. The psychotherapy consisted

primarily of providing a consistent structure to the patient around
which she could reintegrate each time she fragmented.
Patient #6:
She demonstrated the classical mirror transference of Kohut
and suffered from a profound defect in internal soothing mechanisms,

which resulted in her being addicted to holding as described
previously in the literature. The patient varied from time to time in
the form of the mirror transference from an archaic merger to

considering us as twins to an expectation of echo and approval. At
times this was replaced by an idealizing transference, and sometimes
I had the feeling that both were present in oscillation in the same
session, as described by Kohut. This patient, with totally unempathic
parents, suffered from a severe narcissistic personality disorder. She

was able to talk about rage and envy, but it was clear in examining her
relationship with her father that any effort to break up the narcissistic
transferences would have resulted in her experiencing a severe lack

of empathy exactly parallel to what she had experienced from her
parents. There was a serious danger of suicide.
In this case, perhaps the most clearly of all the cases, the
narcissistic problem appeared to be a developmental defect and the

narcissism as it displayed itself in her personality was remarkably
similar to the narcissistic demands of the small child. In this case
there was no essential descriptive difference in the grandiosity of the

patient and the grandiosity of the small child. It is possible that this is
a signal to the therapist, namely, the more his empathic perception of
the patient is that the patient's grandiosity resembles the narcissism of

a small child, the more likely it is that the therapist is dealing with a
narcissistic personality disorder. When narcissistic configurations
approach the bizarre and the delusional, one becomes suspicious that
these are restitutive attempts to deal with unintegrated, sadistic and

aggressive self-and object images that threaten to overwhelm the
patients personality and fragment the self.
Patient #7:
He is an example of the latter. He was apparently not psychotic
but presented a long history of tremendous hostility and even

aggressive acting out, although he never committed any serious
crime. At a point where he was suddenly and unexpectedly
disappointed in a warm mothering relationship with a woman (which
he had unrealistically expected from her), he developed a vivid
delusion that he was the second Christ. This lasted a few weeks and
disappeared and never appeared again, but it shook him up and
caused him to seek therapy. Even with the vivid delusion there was
some insight that it was a delusion.
In this case the narcissistic fantasies were extremely grandiose
as they emerged in the long process of intensive psychotherapy.
Furthermore, they were clearly related to bitter disappointments in

the mothering function and represented the kind of dynamics
described by Kernberg in which the patient split people into good
faces and bad faces; when a person who was a good face failed to
come through, he became instantly transformed into a bad face and

became the object of powerful aggressive fantasies as well as
powerful feelings of fear of retaliation for the aggression. The use of

faces as images in this manner is, in my experience, very common in
the intensive psychotherapy of borderline patients.
In this case the distant, cold and aloof grandiosity was quite
different from that of a child; the patient after a while was able to
discuss his deep, idealizing transference with me as clearly a
projection of his own grandiosity, in which I was to be used as a

stepping stone to his achieving divinity. First he and I would be above
everyone in the world, and then as he used up all my strength, he

would leave even me far behind. There was constant enormous
chronic anxiety over the tremendous aggression attached to these
unintegrated bad self-and object images and a corresponding
superego defect as described by Kernberg, quoted earlier (Chapter
10).
Patient #8:
She presented a cold, aloof grandiose self, bristling with
antagonism, hostility, suspicion and paranoid fears. The pathology
here was that of a starved, lonely, hungry little girl, and the whole
therapy centered around dealing with rage and mistrust of a very
primitive nature. There was clearly a search for a soothing parent, but
no ability to accept any adult for that purpose because of the paranoid
fears. At times I thought a merger transference was appearing, but
whenever the patient got wind that in any way she might need me,
there were immediate denial and paranoid accusations. The problem
was complicated by the fact that her husband supported her paranoid
feelings and her denial and seemed to need her as dependent, sick
and alcoholic. In this case, simply through the therapists becoming a

vehicle for the discharge of her rage and hatred during the therapy
sessions, the patient was enabled to mellow (her term) in her
relationships to other people. At the center of the therapy was a
tremendous fear of rejection and abandonment as well as of

destruction from her projected sadistic preoccupations.
The patient was suffering from a threatened inner
disintegration because of the presence of unbearable highly charged
aggressive affects. Notice that the expression of communication of

these highly charged aggressive affects doesnt have to take place as a
projection of bad or good self-or object images onto the therapist and
their subsequent reintrojection or by projective identification. It can
take place in a number of ways, and it is incumbent on the therapist to
pick up empathically the presence of this volcanic aggressive affect in
the patient and to help the patient to deal with it.
This patient also illustrates another important principle in the
treatment of borderline and narcissistic disorders that has not

sufficiently been emphasized. Simply accepting the patient over long
periods of time, not rejecting him, not getting rid of him, listening to
all the raging, interpreting the projection, and still continuing to

accept and work with the patient often seems to have a very
important therapeutic effect! These patients feel that no one could
possibly accept them or be with them because they know they are so
unpleasant. The great rage makes them feel terribly unlovable, very
anxious and interested in attacking before they are rejected. My
clinical experience has been that the rage these patients present
mellows and calms down just because the therapist does not
disappear, remains consistent and does not retaliate and reject the
patient for it. This is much harder than it sounds, for obvious reasons;
I will discuss countertransference problems in detail later.
Patient #9:
She illustrated similar dynamics to those described by
Kernberg, but in this case the narcissistic aspects were minor and
what was at the center of the defensive structure was a severe

obsessive compulsive set of defenses that appeared after some quasi-
schizophrenic disorganization had been resolved in the
psychotherapy through the provision of consistent structure. The
patient continued to function, but at a severe obsessive compulsive
level, for many years. At the center of her problem was a clear and
profound fear of the hostile sadistic intruding mother, which partly
represented the projection of the patients own aggressions but partly
was also true of the mother.
In this case we see how a different series of defenses can
protect the patient against the same problem as the use of a
pathological grandiosity. In these cases as with many other cases, the

problem is to help the patient deal with overwhelming aggression and
fear of retaliation, as described by Kernberg. This is done in
psychotherapy through the interpretation of transference projections
as well as by providing structure and reality testing for the patient

when the aggressions threaten to fragment the personality. The more
serious the danger of fragmentation, the more the therapist has to

provide himself as a structured introject to help the patient control
the situation. In these cases the patient borrows the strength of the
therapist in the alliance to help buttress the egos defenses against the
disruptive rage.
Patient #10:
A classical type of highly successful narcissist, she came into
treatment when a number of friends who were her late-middle-age
acquaintances died. As a result of this she became aware that ultimate

narcissistic injurysickness and deathcould not be avoided even
for her. The psychotherapy was clearly that of a narcissistic
personality disorder in which a deep archaic merger transference
appeared and sustained the patient for years, resulting in a gross
improvement in her functioning outside the treatment hours.
Gradually there was a transmuting internalization and the patient
became increasingly integrated and healthy, showing clinically the
classical transformation of narcissism described by Kohut. Rage,

when it appeared, was clearly narcissistic rage involving some
narcissistic injury. Profound splitting, described by Kernberg, was not
present.
Patient #11:
She presented herself with identity confusion (her term) and
a terrible problem in choice of career after having tried a whole
variety of occupations. In due time, it became clear that there was a

profound grandiose self demanding to know, to understand and to be
everything that made it impossible for her to be satisfied with any

career choice. In this case an archaic merger transference formed, and
the patient was gradually able to resume her development. At no time
did the profound splitting of object and self-representations occur or
show itself, but rather there clearly appeared to be a developmental

arrest. The case was complicated by an extremely pathological
identification with a schizophrenic father that made the patient
appear much more bizarre and weird than she actually was and much
sicker than she actually was.
Patient #12:
She was almost a converse of the previous patient in that she
presented herself as what appeared to be a classical narcissistic
disorder. Interpretations involving her fear of merger with the

idealized parent imago had no effect, even though the patient seemed
to suffer when such a merger threatened what Kohut has described as

traumatic states. Only when a discussion and working through of
her profound rage took place was the patient able to resolve the
pathology or the phenomenology of the traumatic state and resume

functioning.
Here the patients narcissistic pathology seemed clearly aimed
at protecting her from the frightening emergence of magically

charged archaic aggressive affect. The situation was complicated by
the fact that one of her parents had cancer and was close to death; the
patient felt that the emergence of her archaic rage would literally
cause the death of her parent. This forced a massive repression of

sadistic impulses, which could then only be released in the therapy
process, often by projection. In addition to narcissistic configurations,

the patient had such a problem in dealing with sadistic impulses that
she had to have recourse to a variety of chemicals, including alcohol,
marihuana and others, in order to literally provide a physiological

soothing of the threatening disruptive aggression. It was a good
prognostic sign that the patient was able to give up these chemicals
early in the therapy and focus the problem in the transference.
Her other common defensive technique was withdrawal and

retreat, which sometimes literally meant hiding in her room for days;
at other times it could be experienced within the therapy hour as a
sudden withdrawal of object cathexis from the therapist and a return
to a cold, aloof and grandiose narcissistic state. After a while, I
learned that interpretation of her rage at this point would reverse this
process and allowed the therapy to continue. The serious danger of
suicide was always present in this patient, as one might expect with
so much unneutralized affect.
Patient #13:
She entered therapy with the presenting problem of frigidity,
but it soon became clear that her main aim in life was to render all
males impotent and frustrated and to injure their narcissism in every

possible way. This was clearly a problem of narcissistic rage in which
the attempt was to reverse the real narcissistic injuries she had
suffered at the hands of her father. For example, every time she
reached out as a child to her father for affection she would be teased
and pinched. In this case we were dealing with a borderline patient
with many hysterical features, but the basic psychodynamic

configuration was the tremendous aggression. The narcissistic
problem represented an announcement and an acting out of the
severe aggression, which also manifested itself in many other ways,
such as bowel movements before and after the sessions, overt raging,
constant attacks on and devaluations of the therapist and so on. The
therapeutic problem was clearly to help the patient develop ego
structures to deal with, if not to neutralize, the profound aggressions.
Patient #14:
He was in therapy allegedly because of a depressive
masochistic character disorder. The pathology appeared to be due to
a narcissistic personality disorder as described by Kohut, and the

presence of deep disintegration was not demonstrable. My contention
is that it is possible in many casesalthough from time to time one is

fooledto differentiate between the presence of the narcissistic rage
which is part of any narcissistic personality disorder and the presence
of rage and aggression communicated as infusing archaic

unstructured and unintegrated self-and object images. This
differentiation represents two different types of patients; one suffering
from a developmental arrest in one area of the personality and

another suffering from a defect that formed before the stage of the
cohesive self. The psychotherapist's empathic perception of the quality
of the rage and the quality of the narcissism forms an important source
of information to distinguish between the two types of patients.
Patient #15:
He clearly fell into the borderline type, although he showed no
surface rage at all. He presented with extreme grandiose fantasies

accompanied by fears of world destruction, which were attached to a
series of religious fanatical radio broadcasts. These represented a
carefully guarded secret, and he could not be diagnosed as

schizophrenic on clinical examination! His only relationships were to
inanimate objects, especially cars, radios, television sets, etc. His
secret grandiose fantasies were extremely bizarre and vague, and it

was not difficult to understand that they protected him against the
total fragmentation which would have resulted without them. He was
suffering from an incipient inner disintegration due to total inability
to tolerate extreme rage. This patients example shows that the

patient does not have to present the rage directly in the
psychotherapy for the therapist to catch onto the type of patient he is
dealing with.
Patient #16:
She presents an interesting variation: She broke down when
an ambivalently loved husband was killed in the Vietnam War. This

sudden event confronted her with her own murderous wishes toward
the archaic mother, which she had defended herself against by
complete denialshe loved her mother and she was sure to the
point of delusion that her husband could never be hurt by anybody.
For a long period she could not even bring herself to admit that he
was killed, and there emerged a long-standing grandiose fantasy that

she was a princess of royal birth, which was in some ways fed by (a)
the mothers insistence that her family was superior to ordinary
people and perhaps had some noble blood and (b) her fathers
worship of her as a small child, which suddenly stopped when she
came into the Oedipal period and showed some manifestations of

sexual interest toward him.
Here we have a combination of profound childhood narcissistic
injury with developmental arrest and a problem of apparent archaic
unintegrated self-and object good and bad images in which the
grandiose exhibitionistic fantasies represent both a defense against

the totally destructive rage and a developmental arrest. It was as if
one narcissistic configuration was layered upon another! In these
cases the treatment has to be directed to deal with the profound
pathological splitting so that the patient can function without
fragmentation, and the developmental arrest aspect must be
relegated to a later phase of the therapy.
Patient #17:
There was a similar development of pathological archaic rage,
but narcissistic defenses were not used. The patient used instead
withdrawal from life into a convent and reaction formation by doing
good works. Here again we see how the narcissistic defenses are only
one set of defenses among others that are used by patients against
similar basic problems of dealing with rage and aggression. Breuers
famous patient Anna O. is a similar case (Breuer and Freud 1895).
Finally, it is of the greatest interest to contrast Patients #18
and #19: Both patients presented with the same chief complaints of a
deep sense of being meekly inferior among human beings, depression

and paranoid suspicions. There the similarity ended, for Patients #18
was functionally paralyzed at the beginning of treatment, whereas
Patient #19 was a successful person, well-thought-of, efficient and

married. Both patients presented in due time important narcissistic
and exhibitionistic fantasies, and both patients had been profoundly
narcissistically injured by their parents for different reasons.
Patient #18
He constantly struggled with the problems of fragmentation.
His rage was extreme and his narcissistic preoccupations were
bizarre and total; for example, he wanted to be the dictator of the
world and kill millions of people. He was infused with hatred and
constantly suffered from the projection of this hatred onto others,

including the therapist. Even the most successful interpretations,
when followed by functioning improvement, were often followed by
dreams in which the therapist was murdering him.
Throughout the therapy he stayed faithfully with his
treatment, made new gains, was able to work successfully, get
married and take care of a wife and children, and yet, there was no
change in his basic narcissistic preoccupations or in his inner feelings

of rage. When the rage became overwhelming, he even had to resort
to exhibitionism, which clearly analyzed itself into a hostile attack on
the archaic intrusive mother. The therapy consisted essentially of the

ventilation of his tremendous oral aggression in a nonretaliatory
situation; his destructive fantasies were filled with parallel fears of
retaliation. For example, he could not drive a car, because he was
afraid that if he was stopped for a ticket by the police an altercation

would occur and it would end up with his being destroyed by the
police. Thus the therapy enabled him to function through a ventilation
of the overwhelming rage, but no basic change in the pathology was

possible.
Patient #19
She, on the other hand, formed a merger transference and was
gradually able to reveal important narcissistic exhibitionistic
fantasies from childhood, which were worked through. She showed
the classical oscillation between an idealizing transference and a
merger transference as well as a retreat into cold and aloof
grandiosity when there was a disappointment in the empathic
perception of the therapist. A gradual resolution in her narcissistic
pathology took place, and she showed the classical transformations of
narcissism described by Kohut.
Patient #20
He represents an insufficiently recognized narcissistic
personality disorder that presents clinically with a psychosomatic

disaster consequent to the internalized undifferentiated rage
(Chessick 1977c). The case has been described in detail in Agonie:
Diary of a Twentieth Century Man (Chessick 1977b).
Part IV
PSYCHOTHERAPYGENERAL
APPROACH
Chapter 13
Overview of the Psychotherapy of the Borderline Patient
What are the answers to the unresolved metapsychological

questions raised at the beginning of Chapter 11?
1. Idealization of the therapist can be either a defense and a

projection or a search for the idealized parental imago
or both, depending on the pathology of the patient.
2. The narcissistic configurations may represent a separate

developmental pathway in which there has been an
arrest or, in other patients, may represent desperate
attempts to protect against a fragmenting splitting of
the personality.
3. There is a fundamental difference between borderline patients

and narcissistic personality disorders. In the former,
for the most part, we are dealing with affect clinically
expressed as associated with more primitive and
unintegrated self-and object images, whereas the
narcissistic personality disorder appears clinically to
have progressed farther toward a stage of the
cohesive self.
4. The transferences should be interpreted as defenses only in

those cases where an effort is being made in the
therapy to help a patient in his struggle with
fragmenting rage but should be allowed to develop
without interruption in other cases and in the
intensive psychotherapy of a narcissistic personality
disorder.
5. Kernbergs formulations of the borderline personality

organization are very useful, but they do not apply to
the classical narcissistic personality disorder.
Furthermore, the appearance of good and bad self-
and object representations in the therapy does not
indicate these are present in the infant under two
years of age! At this stage of life there are at most
preverbal, preconceptual, affect-laden memory traces.
When these appear in psychotherapy, they are
attached to representations as they are communicated
by the ego working on a much more advanced level of
cognition and operation. There is no good reason to
postulate the existence of such complex images in the
two-year-old, and most developmental psychology
militates against such postulates. One should keep in
mind the powerful synthetic function of the ego in
telescoping analogous psychological experiences,
especially for the purpose of communication and
being understood.
6. The grandiose self in the borderline patient is crudely different

than that of the child, whereas in the narcissistic
personality disorders it is much more similar and
reflects a developmental arrest.
7. The difference between a borderline personality disorder and a

psychotic seems to have to do primarily with the
tendency toward fragmentation and the cohesiveness
of the ego in terms of coming back together. The
borderline personality fragments to an alarming
degree, but it snaps back rather quickly and without
heroic measures. The psychotic patient fragments into
innumerable pieces; there is a total loss of the
observing ego, and reintegration, when it occurs, is
rigid and brittle. Thus the fragmentation in the
schizophrenic patient is a much more serious and
ominous matter and carries a far worse prognosis.
The psychotherapy of the profoundly schizophrenic patient

has to deal with far greater danger of fragmentation, and only after
the therapist has been built into an important ego structure can the
intensive psychotherapy of the schizophrenic take place. In fact this
may never be possible, because the danger of total fragmentation is so
ominous. It is as if we are dealing with a certain basic glue which is

not present in the schizophrenic, is partially present in a kind of
elastic amount in the borderline personality disorder, but is relatively
strong and cohesive in the narcissistic personality disorder. The

limitation of intensive psychotherapy in these three kinds of
disorders is based on the strength of this glue. Fragmentation always
has to be attended to before anything else, because when it occurs

there is a loss of the observing ego and the fragmented parts of the
personality are not available to the patient for therapeutic work.
Because of this they threaten to disrupt and destroy the treatment in

a malevolent fashion.
The threat of such fragmentation is always ominous and

accompanied by great anxiety in the patient, often along with
psychosomatic symptoms. The alert psychotherapist attends to this
before anything else, and he must constantly assess the danger. In the
schizophrenic patient the danger is always present, and it forms a
profound limitation to the treatment. In the borderline patient the
danger is variable, and careful history taking and experience with the
patient can help the therapist to assess (a) the seriousness of the risk
of fragmentation and (b) the cohesiveness, in terms of how long it will
take for the patient to recover from impending fragmentation.
At the same time one must be careful not to read ones own
anxieties into the fear that the patient will fragment. It is typical in the
psychotherapy of the borderline patient for lots of anxiety to be

stirred up in the therapist, and this sometimes appears as a projected
fear that the patient is on the verge of fragmentation and collapse.
This causes a misjudgment of the psychotherapeutic situation with a

result of backing away from the patient, an avoidance of uncovering,
aborted attempts at support, writing of prescriptions and a
therapeutic stalemate.
In the narcissistic personality disorder the danger of
fragmentation is minimal in a well-conducted psychotherapy and the

closest we get to it are Kohuts traumatic states. These are due to a
flooding of unneutralized narcissistic libido, and if this understanding
is appropriately presented in nontechnical language to the patient,

the excitement usually subsides. As Kohut points out, The analyst
should tell the patient that it is sometimes very hard to become aware
of the intensity of old wishes and needs, that the possibility of their
fulfillment may be more than the patient was able to handle all at
once, and that the present state was an understandable attempt to rid
himself of his excitement ... it can be made clear not only that under
such circumstances the child is in need of a tension-dispelling adult,

but also that the patient is temporarily reexperiencing this old state
since the personality of his mother had not permitted such optimal
experiences in childhood. Eventually the patient with the narcissistic
personality disorder who is experiencing the reemergence of these
powerful feelings in the transference learns to handle these
increasingly familiar tension states more smoothly, without the direct
aid of a therapist.
It is vitally important that the therapist be aware of what is

happening when these states take place during the psychotherapy! If
these states do not respond to such interpretative work, the therapist
should be suspicious that he is dealing instead with the danger of

fragmentation and the disruptive reemergence of archaic oral
aggressions projected onto the therapist as a mother figure. This in
turn serves as an important tip-off to the therapist of the nature of the

basic pathology with which he has to help the patient struggle.
Clearly the basic problem of intensive psychotherapy with

borderline character disorders is how to deal with the fact that the
patient has not received what Winnicott calls good enough holding in
infancy. There are almost as many varieties of recommendations for

the treatment of the borderline patient as there are authors on the
subject. General agreement is only found on a few basic issues. First,
ordinary encouragement or supportive therapy as practiced in the
general physicians office produces either no effect at all or a dramatic

remission soon followed by relapse with the same or new symptoms
accompanied by the angry demand for more magic. Second, the
typical administration of various pharmacological agents to these

patients often complicates the situation in many ways. They abuse the
dosage instructions, and the side effects produced by improper

dosage complicate the symptom picture. They collect medications
from various physicians and take these in various amounts and
combinations. Suicide attempts with these medications pose a
definite risk.
Rapid shifts and changing, with all the excitement, storm, and
panic they cause the patient and those around him, often
accompanied by either missing appointments and failures to pay the
bill or spending session after session in talking about various
symptoms and the constant introduction of extraneous problems and

extraneous matters, soon make both physician and patient feel that
no progress is being made. There is typically an exasperation on the
part of the therapist as well as a developing barrage of complaints

about the treatment from the patient, which may lead to an impasse,
and a referral for hospitalization or a variety of other ways are
employed to get rid of these patients.
However, if one is willing to put up patiently with a great deal
of frustration and disappointment, it is possible to successfully treat

borderline patients. Four basic approaches to the psychotherapy of
the borderline patient are found. The first type emphasizes a very
authoritative and direct approach with much psychological pushing
and shoving of the patient in order to get him moving. Emphasis is

placed on controls, socialization and reality testing, and therapy deals
mainly with the symptoms and attempts to produce a personality
who modifies himself to please the therapist in order to function

socially. Unless interminable contact is maintained with the patient,
relapses are soon to be expected, especially when life stresses arise. If

this approach can be made to work, it is certainly quicker and cheaper
than long-term intensive therapy. An excellent write-up of this
approach may be found in the first edition of the American Handbook
of Psychiatry in an article on the borderline patient by Melita
Schmideberg (1955). It may interest the reader to compare this
article with my article in the second edition of this handbook
(Chessick 1975) to see the advances and the changes in thinking over
a period of fifteen years about our approach to the borderline patient.
The second type of approach is a formal psychoanalysis, seen

by some as a desperate heroic measure and by others as the
treatment of choice. Most psychotherapists reject this approach out of
clinical experience in which many borderline patients show a

complete intolerance to the ordinary psychoanalytic situation,
reacting with suicide attempts, transitory psychoses, dramatic chaotic
symptoms or acting out that finally interrupt the treatment. To say
the least, a formal psychoanalysis of the borderline patient should not

be attempted by anyone except the most experienced and well-
trained psychoanalyst who is willing to assume great risks.
The third type of psychotherapy attempts to combine an

uncovering psychotherapy with providing a directly gratifying
experience of some kind for the patient. This direct experience can
vary from taking the patients hand to examining the patient in the
nude or letting the patient bite and suck on the therapists hand or
breast or what have you, in a direct attempt to provide better

mothering experiences within the particular psychodynamics of the
patient. Needless to say, the danger of massive countertransference

acting out is quite rampant in these situations, and the most hair-
raising and destructive behavior by the therapist can be excused as
attempting to provide a corrective experience. From a theoretical
point of view, this approach has additional dangers in that the use of
such heroic measures, which are essentially a primary-process kind of
interchange, works directly against the stated aim of diverting the
patients ego functioning away from primary-process and toward
secondary-process-based thinking and behavior. Even more than in
authoritative and directive psychotherapy, the patient can easily

become hung up on the primary gratifications involved, leading to a
demand for more and more and at best a subsequent stalemate. I have
seen this repeatedly occur when attempted by inexperienced or

poorly analyzed psychotherapists.
I have collected a series of cases from supervision, from the
reported experiences of other therapists and from patients reports
that I considered reliable, which detail psychotherapists attempts to

treat borderline patients by giving direct gratification of all varieties,
from handholding all the way to hugging and kissing the patient. I
have not seen one single case where this has had any lasting benefit in
the psychotherapy. It sometimes gives the patient a temporary sense
of soothing but invariably leads to a fixation on this kind of

gratification, and alwayswhich is most interestingat some level
the patient develops a profound hatred of the therapist, because the
patient intuitively recognizes (a) that there is an exploitative aspect to
this the therapist is getting some kind of acting-out gratification

and (b) there can be no possible future for this kind of relationship
with the therapistit is bound to end in loss and rejection for the
patient.
This leads to an intense ambivalence in the patient in which he
becomes tremendously hung up on getting the primary-process
gratification from the therapist and at the same time develops a

greater and greater hatred of the therapist. This is usually resolved by
massive acting out outside of the therapy of all sorts and kinds,
sometimes without the therapist even being aware of it. It is very
hard for such patients to leave therapy, because they are so hung up
on the gratification, and when the time comes that the therapist is
sick and tired of hugging the patient or whatever, explosive reactions
occur.
I emphasize this because it happens all too often that the

tremendous anxiety stirred up by borderline patients in therapists is
resolved by massive acting out on the part of the therapist toward the
patient. Again I emphasize that I have never seen a borderline case
where any kind of direct gratification of a patient had a lasting or

useful therapeutic effect. In every instance it has caused more harm
than good. Invariably the reasons given by the therapist for what he is
doing represent rationalizations of acting out.
The treatment offering the greatest potential with the least
serious risk for borderline patients goes under the various names of
psychoanalytically oriented psychotherapy or psychoanalysis with

parameters or, best, psychoanalytically informed psychotherapy. I
will now review this in more detail.
Chapter 14
Psychoanalytically Informed Psychotherapy
The initial problem of psychoanalytically informed

psychotherapy of the borderline patient is getting the patient to form
a therapeutic alliance in spite of all the storm which his symptoms

lend to the relationship. In fact, the patient must at first be very
tightly locked (Chessick 1966) into the therapy to enable him to
maintain the relationship in spite of the terrific anxieties of
abandonment, penetration and annihilation that invariably arise and
must be worked through along with the primitive rage. A very long
period of being there from a psychotherapist with high empathic
capacity and great frustration tolerance is at times necessary before
the patient begins to build a sense of confidence and becomes locked

into a symbiotic relationship with the therapist. This is facilitated by
concentrating at the beginning on reality problems instead of getting

lost in fancy or highly intellectual dream interpretations or
psychodynamic formulations, and also by a certain deep inner
attitude toward his patients on the part of the therapist.
This deep inner attitude is described in a poetic little one-page

paper by the French psychoanalyst Selma Nacht (1969): It seems to
me that what is most important to obtain such a result is not so much
what the analyst says as what he is. It is precisely what he is in the

depths of himselfhis real availability, his receptivity, and his
authentic acceptance of what the other is which gives value,
pungency, and effectiveness to what he says.
Now, of course, this is very very difficult to teach for it is based
partly on ones innate capacities and partly on the thoroughness with

which one has had a psychotherapy of ones own. One must
remember, as Nacht explains, that the activity of the therapist is
aimed at helping man master the incessant whirlpool created by the

clash of constructive and destructive tendencies within him. If he can
manage to escape the zone of conflict, man is able in this way to
escape the ambivalence which constitutes the most pernicious poison

to his psyche, the major obstacle to the blossoming of the forces of
love within him. These forces of love are infinitely more powerful in
man than he may guess, provided that they are no longer constantly
opposed, used or destroyed by conflictual currents.
Thus, as in the treatment of the adolescent, a certain sense of

optimism and confidence in the outcome and in the forces of love
within the patient is necessary if one is to successfully treat
borderline patients. Especially at the beginning of the therapy, this
kind of optimism and confidence helps one withstand the tremendous

vicissitudes which otherwise would make one either retreat into a
passive withdrawal from the patient or surrender in a kind of
masochism to the patients berating activities or even act out by

directly gratifying the patient or by getting rid of him.
If the initial locking in takes place, strong transference
manifestations appear, affording the opportunity to correct or at least
to ameliorate the preverbal disaster without the use of dangerous
heroic measures. This correction takes place in the context of the
transference through empathic understanding and interpretation by
the therapist as well as through a deep emotional interaction between

therapist and patient. Success or failure in treatment depends on this
process.
Various kinds of transferences take place, as I have already

discussed. One sometimes sees Modells type of transitional-object
transference or one sometimes sees Kohuts type of mirror
transference or idealizing transference. As long as these transferences
are workable we have hope for progress. Some of them are

analyzable, resulting in structural change, and some of them are not
analyzable, in which case we get amelioration of the defect and a
certain amount of resignation. We also have to face the fact that there
are limitations as to how much we can do. In general, the literature is
overly optimistic, and tends to ignore the warnings of Anna Freud
(1969) regarding the intensive psychotherapy of pregenital disorders.
When you get into the discussion of the psychotherapy of the

borderline patient, the work of Strupp (1973) becomes especially
interesting and timely and important. Strupp feels that it is arbitrary
in psychotherapy to pin the preeminent therapeutic influence on the
interpretation of the transference. I think there is general agreement

about this as long as one is not talking about formal psychoanalysis.
In the psychotherapy of the borderline patient, except in those cases
that form a transference neurosis that is stable and analyzable, a
variety of factors influence the therapy and they are very nicely
reviewed by Strupp, essentially in terms of learning process. It is
often less important what the therapist considers theoretically to be
causing the change than what the therapist is actually engaging in
with the patient; the latter is often what leads to the therapeutic
change.
Thus the psychotherapy of the borderline patient provides an
excellent area for the research and investigation of therapeutic

influence. Strupp reviews a variety of factors that lead to influencing
the patient in psychotherapy, for example, the importance of a solid,
reliable and trusting relationship with the therapist. They are what
Strupp calls learning experiences in constructive living, that is to say,
in creating a situation where the patient is willing to listen to the
therapist. In overcoming mistrust and resistance to accepting the
therapists guidance, the patient experiences a corrective type of

meaningful experience, and this is important in terms of influencing
the patient to change. There is even what Strupp calls moral suasion,
which is implicit in the therapists apparently neutral clinical stance,
and I will talk more about this in detail later.
Thus the therapist, according to Strupp (1975), establishes

himself as a good parent or authority figure: . . . he creates conditions
that maximize the chance of his being listened to and he seeks to
neutralize or undercut road blocks the patient places in the way of his
teachings; he points out maladaptive patterns of behavior and their
underlying infantile assumptions; he sets an example by remaining

calm, unruffled, reasonable, and rational; he refuses to get entangled
in the patients neurotic machinations; he conveys the message that
the patient must learn to accept personal responsibility for his own
actions instead of blaming others and life circumstances for his
predicament; he teaches the basic lessons on how people in Western
civilized society interact productively and nonneurotically; he teaches
the patient to be less demanding and grandiose, to scale down his
expectations of others, and to accept a more active role in managing
his life; he conveys a philosophy of reasonableness, rationality,

moderation, mutuality and fairness as the guideposts of the good life;
and, in broad terms, he combines love with discipline in helping the
patient become a more autonomous, self-directing, and responsible

adult.
This becomes most controversial when one is talking about
classical psychoanalysis, in which the analysis of the transference
neurosis makes the major changes, but in psychoanalytically

informed psychotherapy it is not necessary to do what Strupp calls
smuggling in these lessons on constructive living through the back

door! We know that these are important therapeutic influences in
psychoanalytically informed psychotherapy, and if one is to be
successful in bringing about the very difficult task of inducing anyone

to change deeply ingrained habits and beliefs, one is going to have to
present ones self (Havens 1974) by involving the patient in an
emotionally charged relationship and utilize the dependency and the
transference to influence the patient in desired directions.
One of the most theoretically important questions is whether

if one wishes to bring about basic structural changethis is enough.
Does there not also have to be some semblance of a transference
neurosis, which can be analyzed in order to resolve infantile conflict?
Assuming that we are sufficiently analyzed ourselves and are
not directly influencing what is happening in terms of transference
formation, we don't have much choice as to whether a transference or

a transference neurosis will form or as to what kind of a transference
will form when we are working with borderline patients. Certain
types of transferences are not desirable and they threaten to destroy

the therapy entirely. The transference manifestations in general can
be extremely frightening and strong and the patient resorts to many
unusual measures to deal with them.
Perhaps the most dangerous problem comes from acting out in
the transference relationship. Greenacre (1963) describes a type of

massive acting out in the transference that is frequent, repetitive and
sometimes lasts over a considerable period of time. In her cases it
mainly occurs in the analytic relationship and in the analytic hour, but
she mentions that for some patients it may even extend into
relationships outside of the immediate contact with the analyst. This
acting out usually occurs in the form of attacks, but not invariably in
the form of attacks, and consists primarily of provocative or seductive

behavior, taking many forms.
For example, the patients may present themselves as suffering
and mistreated, with a constant worrying about some current
grievance which even has a core of a kernel of truth in it, but which
becomes brooded over in an obsessional way with a quiet drama
aimed at getting the therapist to make an emotional response or
intercede on behalf of the patient. More frequently the other side of
the coin is up and the therapist is represented as the
misunderstander. Then there is persistent, insidious nagging
accusation with taunting ridicule. Anything the therapist says is taken

out of context and distorted; if the therapist says nothing this is
interpreted provocatively as indicating the justification of the
complaint. There is a trying out of the therapist by a wearing-down

effect to see where the limit of his tolerance really is.
This whole performance has the form of a tantrum of a special
kind in which there is a relentless demand for reciprocation and
discharge through the therapist. Greenacre interprets this as being

connected to projective identification and implies that there is a
beating fantasy behind this provocativeness, but she doesnt discuss it
at any length. The psychodynamics may be controversial and may

vary, but the therapist doesnt forget this kind of experience very
easily.
One must be constantly aware of the potential for such attacks.
Sometimes they occur outside the therapy, so the therapist must keep
a constant eye on what is going on in the patients real life or he will
be suddenly confronted with explosive developments in the therapy
of the borderline patient.
Two such typical dangerous transference developments seen
in the psychotherapy of the borderline patient are the erotized
transference and the involvement of a third person in the

transference, developments both of which must be quickly recognized
and dealt with or the treatment will be ruined. The erotized
transference, which was even recognized by Freud (1915) early in the
development of psychoanalysis, manifests itself by the stormy
demand for genital contact with the therapist and so on. When this is
rejected, the patient experiences deep and sincere hurt and
humiliation. It is not amenable to interpretations and persists as a
demand for direct gratification. Empathy, consistency of approach,

patience, understanding the patients sense of rejection and not
reacting with fear or hostility to his or her demands can eventually
lead to a resolution of this problem. It is very important in such

situations not to make the patient feel humiliated or put down. If the
therapist does not recognize the genuineness of these feelings and

makes light of them or ridicules them, it is experienced by the patient
as a profound narcissistic blow and generates a situation from which
the therapy itself can never recover.
Similarly, borderline patients often cannot stand the intensity
of their longings for the therapist in the transference and they may
quickly dump all of this onto a third person and engage in massive
acting out. In addition to that, they tend to sexualize these longings,
which of course are really much more primitive and pregenital, but
they are more acceptable if they are sexualized and the chances of
acting them out are better. If this is not recognized and interpreted
and stopped, sometimes emphatically, situations such as impulsive
marriage or pregnancy may result.
A word is in order here about how one goes about
emphatically stopping such massive acting out. It is always presented
to the patient in terms of, This will be a danger to you or to your
psychotherapy. It is not in your best interest. It is never presented as
a moral command of any sort, for the major responsibility the
therapist has in preventing acting out has to do with the preservation
of the psychotherapy itself. This rule is only violated in situations
where the patient is a physical threat to himself or others, in which
case the therapist must feel that his duty as physician or citizen comes
first, but most of the time the problem is to emphatically present to
the patient the idea that he is ruining his own psychotherapy by
acting-out behavior and that he must delay it or the psychotherapy

will be destroyed.
If this acting out persists and the patient insists in going ahead
with plans that are clearly involved with transference acting out, the
therapist must confront the patient with the choice of either stopping
his behavior or foregoing treatment until he is willing to control
himself better. In no way should the therapist implicitly condone or

accept behavior on the part of the patient which he knows is a
massive acting out in the transference. Everything else in the therapy
has to come to a halt while this problem is dealt with, as it is a very

dangerous situation for the patient and his psychotherapy. Alertness
to the potential problem and constant active concentration on the
patients life situation are necessary. The use of a third person is not
always undesirable to help the patient withstand the tense

transference longingsit depends on what extremes the patient has
to go to with this third person. Too energetic interpretations of
transference longings can throw a patient into a chaotic panic and

also disrupt the treatment entirely.
If disruption does not occur, the transference is properly
understood and interpreted, and anxieties are gradually worked
through, the patient is often able to uncover grandiose core fantasies

in a protective atmosphere. Borderline patients live around their
early infantile narcissistic fantasies which permeate and contaminate

all ego operations. They cling tenaciously to these narcissistic
fantasies, which represent a consolation for the deprivation of affect
from the mother and also the patients typically attempt to produce
the longed-for affect through satisfying all confusing, conflicting and
unrealistic parental expectations. The patient lives as if he had
already secretly accomplished these fantasies, producing a set of

poorly adaptive responses to life.
Sometimes these grandiose core fantasies are apparent even at
the beginning of treatment, but direct assault upon them simply
results in vigorous denial or even breakup of the treatment, since they

represent substitutes for gratifying human relationships and they
cannot be given up until the annihilation and abandonment fears are

worked through in the transference. A sarcastic approach to such
fantasies, which is a typical beginners mistake, always represents
countertransference difficulties.
It is clear that the basic factor in successful psychotherapy of

borderline patients has to do with how the psychotherapist handles
the crucial dilemma produced by the intense transference longings
and also by the associated deep fears and rage. Frosch (1971) points
out the thin line the therapist has to walk between the gratification of
the patients wishes and the imposition of limits. One must bear in
mind that the borderline patient himself frequently is caught on the
horns of a dilemma, namely his need and wish for the objectin this
instance, the therapistand his fear of engulfment if such wishes are
gratified. The whole problem of giving and receiving permeates the
patients relationship with his parents and subsequently with his
therapist.
In agreement with what I have written, Frosch essentially

points out that this very state requires structuring and the imposition
of limits to help the patient deal with the multitude of confusing
factors that make it so difficult to structure for himself, and he

explains that it is possible to do this more firmly after the distrust
which frequently permeates the feelings of the borderline patient for
the therapist has been diminished and a good psychotherapeutic
relationship has begun to evolve. Thus the building up of a feeling of

trust is the crucial aim of the earliest steps in the psychotherapy of
the borderline patient.
In a paper (Chessick 1968) written before that of Frosch, I

pointed out how the therapist has to walk a tightrope in this crucial
dilemma. On the one hand, it is clear that direct ministering to the

patient's needs by behavior such as caressing or feeding or giving
gifts to the patient constitutes a form of acting in, as I have
described (Chessick 1974a). It is undesirable except in the most

minor and socially acceptable forms, such as allowing a cigarette to be
borrowed and so on, because it prevents ego expansion by fixing the

patient on the omnipotence of the therapist. On the other hand, a
therapy without parameters will not hold the patient in treatment.
The crucial dilemma the therapist always faces has to do with the
question of where to draw the line.
For example, watching neophyte therapists it is easy to show
that hiding behind rigid adherence to technique or rules of treatment

is a defense against feeling the anxiety engendered in them by the
massive pregenital strivings of borderline patients. Analogous,
perhaps, is societys tendency to treat such people with rigid rules, for
example, The army will make a man out of him. On the other hand,
dangers to intuitively approaching such patients are obvious. The
therapist must genuinely know himself and not engage in

countertransference acting out. The patients are only too eager to act
out or act in in the therapy, and they pose a threat to the neophyte
from that point of view alone. Thus, the key factor behind
improvement is the therapists ability to be emotionally responsive

without predominantly using the patient for his own needs. In a
similar fashion the therapist must be able to draw away and permit
separation and individuation at the proper time.
Obviously, this depends primarily on the self-understanding of
the therapist. Careful study of case material shows that it is actually

possible to keep a secondary-process check on what is going on so as
to avoid a wild and disorganized therapy. The more thoroughly
understood the patient is, the more accurately it is possible to know

whether our emotional interaction with him is on the beam from
session to session. Improvement in the patient appears to be directly

related to this emotional interaction, and to the degree to which it is
consistently genuine, on the beam, and originates from a healthy and
positive deep inner attitude of the therapist.
Chapter 15
Therapeutic Regression
The task of psychotherapy with psychotic and borderline

patients becomes infinitely complicated by the fact that the patient
neither has a firm grasp of his own sense of identity nor is able,
because of his primitive narcissism or his need to project malevolent
introjects, to respond to supportive kindly or benevolent measures as
we would logically expect a starved and lonely person to do. It is now
theoretically clear why attempts to directly gratify the borderline
patient repeatedly fail. What is necessary instead is for the therapist

to empathically grasp the nonverbal, unconceptualized affect-laden
memory traces as they are communicated in subliminal inflections
and behavior, to call attention to these and to help the patient

conceptualize and verbalize them and explore their origins and
meaning. A long period of working at this primitive level of

education is often necessary before any interpretations make any
sense to the patient.
We must begin at the level of ego function and perception the
patient is at and gradually enable him to move forward

developmentally to a level of cognitive and intellectual function
where thought and abstraction make sense at all. This is the secret of
the frequent impulsiveness of the borderline patient; it is not

malevolencethere is simply no thought barrier developed between
the impulse and the act. First of all we must literally help the patient
bit by bit, through focus on his behavior, to develop a stronger and
stronger thought barrier and capacity to wait and delay action.
The key to any successful uncovering psychotherapy with the

borderline patient is in the capacity of the therapist to permit and
enable the patient to unfreeze disruptive and restrictive introjects
warping the basis of his early ego function, a function of the

therapeutic atmosphere the therapist creates for all his patients. The
major source of resistance to psychotherapy, as pointed out by
Guntrip (1968), is . . the extreme tenacity of our libidinal attachments

to parents whatever they are like. This state of affairs is perpetuated
by repression in the unconscious inner world, where they remain as
subtly all-pervasive bad figures generating a restrictive, oppressive,
persecutory, inhibiting family environment in which the child cannot

find his real self, yet from which he has no means of escape.
The only reasonable approach to these patients in uncovering

psychotherapy will have to be in ultimately unfreezing the early ego
formation, an unfreezing which can occur only if a controlled
regression is permitted to take place. It must be pointed out that

regression cannot be forced by the therapist. It must occur as a
natural consequence of the sense of security within the therapeutic
alliance that is allowed to form between a relatively healthy therapist

and whatever mature aspects are available in the observing ego of the
patient. Winnicott (1958) enumerates some of the obvious factors

that allow this regression to take place. It might be first mentioned
that the whole thing adds up to the fact that the analyst behaves
himself or herself, and behaves without too much cost simply because
of being a relatively mature person.
The factors encouraging a regression useful in psychotherapy
are
1. A consistent and frequent being at the service of the patient, at

a time arranged to suit mutual convenience
2. Being reliably there, on time, alive, breathing
3. For a limited period of time, keeping awake and becoming

preoccupied with the patient
4. The expression of love by the positive interest taken and hate

in the strict start and finish and in the matter of fees
5. The attempt to get into touch with the process of the patient, to
understand the material presented and to
communicate this understanding by interpretation
6. Use of a method stressing a nonanxious approach of object

observation
7. Work done in a room that is quiet and not liable to sudden

unpredictable sounds and yet not dead quiet; proper
lighting of a room, not by a light staring in the face and
not by a variable light (In some instances the patient
lies on a couch (Chessick 1971b) that is comfortable in
most instances, depending on the situation, a face-to-
face situation with the patient is necessary)
8. Keeping moral judgment out of the relationship as well as any

uncontrollable need on the part of the therapist to
introduce details of his personal life and ideas
9. Staying, on the whole, punctual, free from tempter tantrums,
free from compulsive falling in love and so on, and in
general neither hostile and retaliatory nor exploitative
towards the patient
10. Maintaining a consistent, clear distinction between fact and

fantasy, so that the therapist is not hurt or offended by
an aggressive dream or fantasy; in general eliminating
any talion reaction and insuring that both the
therapist and the patient consistently survive their
interaction (Winnicott feels that this setting
reproduces the earliest mothering techniques and
invites regression. If it is consistently offered, an
unfreezing takes place as a natural consequence of the
regression that occurs)
The crucial unfreezing of malevolent introjects through a
controlled regression contains within it two major constructive and
therapeutic events. The first of these is the loss of destructive

introjects; the second is the substitute introjection of the psychic field
offered by the therapist.
However, the regression also contains a major potentially
destructive event, for such a regression stirs up omnipotent
expectations on the part of the patient; a yearning for what the
therapist can do, magically and in a primary-process manner, to

restore to the patient all the missing experiences from his infancy and
to make good for the patient all the negative experiences of his
infancy. I shall proceed to discuss first this potentially destructive

event and then the constructive events in detail.
Either the inevitable frustration of the omnipotent
expectations of the patient stirred up by regression or narcissistic
blows that occur to the patient in real life during the long process of
psychotherapy can lead to a series of events that result in a failure of
the treatment. The psychotherapist must be aware that the danger of

regression induced by the therapeutic setting can lead to failure, and
he must have an understanding of the typical kinds of consequences

that are produced as a result of the frustration of the omnipotent
expectations and from serious narcissistic blows.
Such consequences are typically:
1. Acting out, in which the patient quits the treatment or in which

he quickly finds a third person to meet his unbearable
infantile cravings for holding and body contact as
described by Hollender (1970; Hollender et al. 1969,
1970)
2. The need for revenge, in which the patient through passive

aggression stalemates the treatment, stalemates his
life or allows his life situation to fail, making
psychotherapy impossible
3. Projection of destructive introjects onto the therapist with fear

and hatred of him, all coming as a consequence of the
frustration of the patients omnipotent expectations,
which may lead to a breakup of the therapy
4. An autistic retreat on the part of the patient into sadistic sexual

fantasies
5. At worst, hallucinations and delusions as a consequence of self

fragmentation, which may even require
hospitalization
Let us turn now to the therapeutic and constructive
consequences of regression in the treatment setting. The loss of
malevolent introjects as a consequence of regression can be noted if

the therapist carefully studies changes in both ego function and
superego function in the patient. Destructive introjects in the ego

manifest themselves by poor adaptative techniques. Introjects of the
parents include many elements of the relationship with them that
involve methods of mastery. As there is a loss of destructive and

restrictive introjects, there is a corresponding improvement in the
patients capacity to adapt to the external world, to function more
efficiently and to observe himself more realistically. Similarly, as

introjects are discharged from the superego by projection and then
interpreted, there is a lessening of the hostile punitive aspects of the
superego, and the patient becomes more reasonable with himself and
others and begins to develop a sense of beloved self.
Our basic tool in intensive uncovering psychotherapy is
introspection and empathy. Kohut (1959) explains that consistent

introspection in the narcissistic disorders and the borderline states
leads to the recognition of an unstructured psyche struggling to
maintain contact with an archaic object or to keep up the tenuous

separation from it. In borderline states, archaic interpersonal
conflicts occupy a central position of strategic importance that
corresponds to the place of structural conflicts in the

psychoneuroses. Thus Kohut points out that the analyst is
introspectively experienced within the framework of an archaic
interpersonal relationship. He is the old object with which the

analysand tries to maintain contact, from which he tries to separate
his own identity, or from which he attempts to derive a modicum of
internal structure.
Giovacchini (1965) points out that the acquisition of adaptive
techniques by the ego leads to a higher state of integration and
involves the process of incorporation. The ego develops by acquiring

introjects that lead to more efficient functioning. He continues, Such
introjects may have several modes of action. On the one hand, the
introjects may act defensively, reducing the disruptive potential of
intrapsychic conflict or ego defect and permitting the ego to achieve a

more stable homeostasis; as a consequence areas of functional
autonomy may develop. On the other hand, the ego may utilize an
introjected positive experience not only in regard to its defensive

potential; it may benefit from the experience directly by having
assimilated an adaptive technique. . . . The egos armamentarium is
expanded and its functional range is increased.
Now when the desired controlled regression takes place in

psychotherapy, a dissolution takes place, giving the ego the capacity
to incorporate new objects, a capacity it did not previously have
because of hostile destructive introjects leading to constriction. By
regressing to such a level of disorganization, the ego has also lost its
capacity to maintain the structured introjects when it progresses into
a slightly advanced position. It has gained from the loss of such
introjects insofar as it has the capacity to incorporate experiences
which can expand its adaptive potential.
Insofar as the child experienced an assaultive and rejecting
external world before there was self-object differentiation, adult
levels of ego functioning will reflect disturbances in structure instead
of the id-ego conflict of the psychoneurotic. On this point there
generally seems to be considerable agreement. Any attempt on the

part of the therapist to directly gratify the patient will be offering the
incorporation of an interaction that is based on primary process and
thus cannot become a basis for ego development. If we attempt to

give primary-process gratification by responding to irrational
demands, we are contributing to the maintenance of an infantile
organization and an equilibrium which contained a preponderance of

primary-process elements. As Giovacchini points out, the
therapeutically desired development and synthesis always heads in

the direction of the secondary process.
To put it another way, we want to do anything we can to

enhance the egos drive towards autonomy. In a later paper
Giovacchini (1967a) discusses the need for acting out and what he
calls externalization. The patient cannot tolerate the helpfulness of
the therapist. By providing a setting to facilitate regression, the

analyst situation sometimes causes the patient to believe that
gratification is possible and reinforces the expectation of primitive
satisfaction. The patient hopes to be rescued from his assaultive and

depriving introjects, a megalomanic expectation of rescue which leads
to bitter disappointment. As a result the therapist is viewed as
insincere and is converted into a replica of the frustrating
environment the patient once knew. This transformation involves a

projection of a bad self as well as externalization; I have discussed
externalization in detail elsewhere (Chessick 1972b, and in Chapter
20).
Giovacchini (1973) conceptualizes the borderline state as one

in which the patient lacks adaptational techniques because he lacks
memories of early gratifying experiences that later should have
developed into methods of dealing with the problems of the outside

world. The functional introjects which later contribute to the egos
executive techniques of mastery were never formed or, relatively
speaking, only imperfectly formed. Giovacchini points out how these
patients often create a situation of tension and urgency by expressing

a need to be helped which cannot be met, since neither the patient nor
the therapist knows what kind of help the patient seeks. The needs
are so primitive that they cannot be articulated, and the therapist, if

he tries to respond directly to them, experiences the same frustration
and helplessness as the patient.
Giovacchini feels that the diagnostic evaluation of borderline
patients is best made on a combination of various qualities of ego

structures, behavioral characteristics, and possible courses of the
disorder, a combination which also takes into consideration
therapeutic outcome. I am in agreement with this approach.
Giovacchini concentrates on ego systems, because he is interested in

the formation of ego systems through the process of introjection. He
divides these ego systems into three general categoriesperceptual,
integrative and executivebut states the behavior and adjustment of
the patient considered borderline indicate that the primary defects

are in the integrative system of the ego. The integrative system
coordinates perceptual stimuli, either inner needs or demands from
the outer world, with appropriate executive responses. These
appropriate responses depend upon memories of past satisfying
experiences. If such experiences are lacking one does not know how
to respond.
The lack of functional introjects in the borderline patient
further leads to constrictions in the executive system of the ego,
although the perceptual and integrative systems are also involved.

This leads to a conflict between the ego and the demands of the outer
world and to a tendency, because of this, for the patient to withdraw
and have tremendous feelings of vulnerability.
Speaking of the symbiotic stage, Giovacchini continues, One
need not conceptualize this stage in terms of subtle mental

representations, since the infant has neither the emotional nor the
neurological structure for complicated mentation. In terms of needs
that is, a reestablishment of homeostatic equilibriumhe need not

distinguish between himself and the person who administers his
needs. Emergence from the symbiotic phase results in the
establishment of a structured identity so that eventually boundaries
between the inner and outer world are clearly established. Insofar as
the borderline patient has suffered frustration and deprivation during
the stage of symbiotic fusion, his self-image is imperfectly formed.
Since he has received little gratification, he has very few adaptational

techniques to cope with even pedestrian problems. Note here that it
is during the symbiotic phase and the vicissitudes that take place
during it that Giovacchini places the beginning development of the
borderline state.
Thus the borderline patients early transference represents an
effort to experience a symbiotic relationship with a strong, powerful
person who will supply him with the adaptive techniques denied him
by an inadequate, weak mother. The patient hopes to experience a
magical rebirth and be the master of the now-threatening outer

world. How the borderline patient reacts to the disappointment of
magical expectation is a very critical issue. Giovacchini (1973) feels it
distinguishes the borderline from other patients suffering from

characterological problems, and he sees the typical reaction as
poignant sadness, with the patient believing that he simply cannot be
helped by treatment.
The reaction to the patients frustration is counterfrustration
in the therapist. This often leads the therapist to become anxious,
because he has identified with the patients desperation, feeling that
he must respond almost blindly, giving advice or management. At this

point the therapist abandons his analytic role because of his sense of
urgency and anxiety. Sometimes the patient accepts this

abandonment of the analytic role, but more often he feels that such
intervention is an intrusion, or at least he eventually feels frustrated
by it, since it is not really an appropriate response. It is felt more as an

assaultive foreign body, and in Giovacchinis experience and also my
own, it often stirs up a tremendous amount of rage which may not show
itself until much later.
Many treatment failures marked by the patients sudden

withdrawal from treatment are caused by the therapists anxiety.
Giovacchini points out that the therapist attempts to defend himself
against anxiety, by being professional, which may result in forced
interpretations. The situation becomes unmanageable and filled with
quarrels and defensiveness. Then both patient and therapist feel
helpless and therapy becomes increasingly confused and chaotic.
Since the patient feels even more helpless than usual, demands for
rescue increase and the therapist feels even greater anxiety. A vicious
circle is created with an atmosphere of frustration and

counterfrustration.
It must be made clear that if the therapist, on the other hand,
refuses to abandon the analytic attitude and observational viewpoint,

he may cause unmanageable amounts of frustration and run the risk
of losing the patient. If the patient terminates treatment under these
circumstances, it is doubtful whether intensive psychotherapy can
help him. Patients sometimes feel they need something more than
psychotherapy, and when the therapist refuses to modify the
treatment and begin gratifying them, they leave. In my experience this

leaving is either poignant or filled with rage and recrimination;
either way it constitutes a sincerely painful experience for both
therapist and patient.
More is internalized from the therapist in a properly
conducted treatment than the healthy experience of a correct
interpretation. First of all, the therapists nonanxious observing

attitude, his compassionate, studious and sincere approach to the
patient, becomes a part of a healthy introject in the patients ego. Most

tricky of all, it seems imperative that we recognize the
countertransference structure (Tower 1956) as an important aspect
of the therapists attitude, forming a psychic field that is also
introjected by the patient. If this countertransference structure is not,
for the most part, malevolent and murderous, at least it does not
represent a serious impediment to ego growth through introjection of
the therapists psychic field. We hope then, that through the
introjection of the psychic field of the therapistbased on correct
interpretations, a compassionate secondary-process approach, a

nonanxious observing attitude and a relatively benign
countertransference structurethere is ego growth, manifested by
better functioning of the various subsystems of the ego.
This view is supported by research at the Menninger
Foundation (Appelbaum 1975) in which internalization is viewed as
made up in long-term psychotherapy of a number of part processes
that are the result of a growing working relationship and that act in a
reciprocally enhancing manner with the therapeutic relationship,
contributing to its development. These interrelated part processes

are: (1) the enhancement of self-esteem; (2) the corrective emotional
experience; (3) transference curei.e., changed behavior in order to
impress the therapist; and (4) identification with the therapists

attitudes. The Menninger group believes that all of the mechanisms
which are conceived of in Kernbergs terminologyproduce
structural change, which becomes reinforced by the environments
favorable response and so becomes increasingly stable. The surprise

in the research was the extent of change that occurred in patients
unable to utilize insight. This again emphasizes the crucial
importance of the actual experience the patient has with the therapist
in long-term intensive psychotherapy.
Chapter 16
The Psychic Field of the Psychotherapist
It is ego growth through this process that allows the

destructive dangers of the regression induced by the therapeutic
setting to be overcome by the patients gradually increasing ego

strength. They will be overcome providing certain ominous situations
do not occur:
1. If the omnipotent demands are not too overwhelming and

immediate
2. If the patient does not immediately quit or unconsciously set

out to destroy therapy before any work can be done
3. If the destructive introjects that have made up the early ego

formation of the patient are not so constricting and
malevolent that a total rigidity and incapacity to get
free of them is present
4. If the psychic field of the therapist is mature enough
It is clear that a certain realistic limitation is placed on our
therapeutic efforts by the first three of these factors, and some cases
will inevitably fail because of them. It is in the area of the psychic field
of the therapist that the most hope exists for an improvement of our
results.
In Why Psychotherapists Fail (Chessick 1971c) I have already
discussed in detail the production of the optimal psychic field of the
therapist. In addition to this generally optimal psychic field, there is

certain specific work that the therapist must do with every patient
and in every psychotherapy. It is easy to see that a

countertransference structure is stirred up in the therapist (a) by
each regressive step in the patient, confronting the therapist with a
new set of feelings, demands and reactions; (b) by intercurrent

realistic or narcissistic blows in the life of the psychotherapistafter
all, this is long-term psychotherapy, during which both therapist and
patient are experiencing numerous events in their actual living; and

(c) by the very length of time of therapy, representing a time
frustration to the secret omnipotent hopes of the psychotherapist.
All these factors operate to provoke the tendency in the
therapist to exploit or retaliate or both, even in such minor ways as

the tone of his voice or letting the patient out a minute early. Thus a
constant self-analysis of the countertransference structure must be

going on within the therapist in order to keep the psychic field up to a
maximum of maturity. This should take place at the same time as
efforts are made to understand the patient and to interpret this

understanding back to him. So, in the language of scientific
understanding, learning from ones patients means expanding ones
own ego capacities through the continual self-analysis of

countertransference structures precipitated either by the various
phases of the patients regression or by intercurrent events in the real
world of the patient and/or the therapist.
This is not enough. In the language of the humanistic
imagination, the patient must continuously experience the presence
of the therapist. Each therapy session must count, as Saul (1958)

puts it. Each session must represent an encounter between the
psychic field of the therapistwhich in its maturity, extends trust,

confidence and hopeand the need-fear dilemma of the patient, who
has fallen away from authentic living and being with another person.
This deep inner attitude on the part of the therapist can be

maintained only as a function of continual reduction of constricting
countertransference structures, just as a healthy nervous system
permits the maintenance of an alert and attentive mind. The purpose

of this for the patient is described by Saul (1970):
For the unsustained, the analyst must provide the

experience which the patient lacked in childhood:
that of having an interested, sympathetic,
understanding person always available in his life.
Without such an attitude, technically correct
interpretations may be interpreted by the patient as
disapproval. Accurate interpretations also require an
attitude of human understanding, of being on the
patients side, of having confidence in him. . . . The
analysts confidence is partly internalized and can
move even the hollow ones in the direction of a
sense of sustainment, of identity, a good self-image
and self-acceptance.
It follows from these theoretical considerations that the

phenomena described in the language of the humanistic imagination
such as presence or being there or I-thou and so on represent
epiphenomena of the successful working through in the
psychotherapist of the various phases of countertransference
structure called forth by the phases of the patients psychotherapy.
The reverse is not true; here is where an increasingly common
amateur error takes place. One cannot force presence or an

encounter down a patients throat. Hugging and touching and going
through all kinds of immediacy gestures with a patient will not fool
the patient; they mask serious countertransference problems in the

therapist. Only the natural and inevitable unfolding of a human
encounter in the forward progress of psychotherapy, as both the
patient and therapist work through their respective tasks, can

produce a genuine growth experience for both. There are no shortcuts.
Balint (1968) insists that in some cases in which wordsthat

is, associations followed by interpretationsdo not seem to be able
to induce or maintain the necessary changes, additional therapeutic
agents should be considered: In my opinion, the most important of

these is to help the patient to develop a primitive relationship in the
analytic situation corresponding to his compulsive pattern and
maintain it in undisturbed peace until he can discover the possibility
of new forms of object relationship, experience them, and experiment

with them.
According to Balint, the task of the psychotherapist with
patients who are not classical neurotics, but whose disorders have
begun before the consolidation of the repression barrier, is essentially

to supply a new beginning to the patient. He attempts to provide an
atmosphere in the psychotherapy that in a sense is a corrective
emotional experience to the early nonempathic mothering the patient
had. Those who follow Balint emphasize the patients absolute need
for empathy from the therapist and stress the danger that
inappropriate verbal interpretations may be produced, because it is

the empathic interactions that are essential for the successful
treatment of such patients rather than interpretations of a
transference.
Balint gives a few examples of how the unobtrusive analyst
can foster this process. For example, the more the analyst can reduce
the inequality between his patient and himself, the better are the
chances of a benign form of regression. The analyst also provides time

and a milieu that has a holding or therapeutic function. The
environment should be quiet, peaceful, safe, and unobtrusive ... it
should be there and ... it should be favorable to the subject, but. .. the
subject should be in no way obliged to take notice, to acknowledge, or
to be concerned about it.
Again, Balint warns us that by providing this special
therapeutic relationship the analyst must avoid becoming an

omniscient and omnipotent object, and he must be sure that the
gratification will result not in a further increase or excitement in the
patient, but in the establishment of a tranquil, quiet well-being and in
a better safer understanding between the patient and himself. He

adds, None of the details of the therapeutic attitude outlined here are
essentially different from what the analyst adopts when dealing with
patients at the Oedipal level, and even the topics worked with are
usually the same; but there is a difference, which is more a difference
of atmosphere, of mood.
Fundamentally, I do not think that there is any necessary
connection between Balints (1953, 1968) theoretical conceptions
and what he is advising the clinical therapist to do. If one carefully

follows his recommendations, one is not carrying out an active
therapy in the sense of Ferenczi at all, but simply permitting and
tolerating a controlled regression of importance during the

psychotherapeutic process. The therapist is simply being a decent
human being who understands when to push the patient with
interpretations and when to allow the patient some time for peace,
quiet and working through. I do not think that this differs in clinical
practice from how any sensitive, humane and feeling physician would
act with his patient. Thus, there is a substantial gap between Balints
highly controversial theoretical considerations and the general office

practice of psychoanalytically oriented psychotherapy. The difference
seems to be an emphasis on a theoretical explanation of what is and is

not important in therapeutic process rather than any fundamentally
different approach to the patient.
Kernberg (1972b) brings to our attention that Balint sharply

criticized the Kleinian use of conventional language mixed with nouns
(like breast, milk, inside of the body) the meaning of which became so
extended and comprehensive from Kleins clinical work, and he

suggested that the Kleinian analysts tend to develop what he called a
mad language. As reported at scientific meetings and in the

literature, these kinds of interpretations create the impression of a
confident, knowledgeable and perhaps even overwhelming analyst.
He wondered if this might be the reason why there is so much
aggressiveness, envy, and hatred in their patients material and so
much concern about introjection and idealization, since these are
common defense mechanisms in a partnership between an
oppressed, weak person and an overwhelming powerful one. This
deserves some careful thought in terms of psychotherapeutic
technique. When powerful rage reactions show themselves in the

psychotherapy of our patients, we must be extremely careful that they
may be related to the therapeutic technique, language or
countertransference structure rather than simply to the projection of

primitive bad self-object representations.
Kahn (1974) refers to Balints contention that there are really
two styles of relating to the patient:
1. Listening to the patient, to what he verbally communicates in

the classical situation, and deciphering its meaning in
terms of structural conflict (ego, id and superego) and
through its transference expression in the analytic
situation.
2. What Kahn calls a psychic, affective and environmental

holding of the person of the patient in the clinical
situation. The alleged result of this style is to
facilitate certain experiences that I cannot anticipate
or program, any more than the patient can. When
these actualize, they are surprising, both for the
patient and for me, and release quite unexpected new
processes in the patient.
My big objection to this second style is that it is too mystical.
Even Kahn agrees that it is very personal to the style of living of the
patient and of working of the therapist; as such, it becomes very
difficult to teach and can be used or abused as license to do just about

anything to and with the patient that the therapist wishes to do.
Balint, of course, is extremely careful to avoid this sense of
license. He distinguishes between two important types of regressions,
which he calls malignant and benign regression. Regression for the

sake of gratification, which has the qualities of despair and passion
and aims at gratification by external action with a suspiciously high
intensity of demands and needs, is Balints conception of malignant
regression.
He sharply distinguishes this from a regression in which what
the patient needs is the arglos state. What is desired in this state is
the analysts recognition of the patients needs and longings for
satisfaction which are the essence of a new beginning and the

patients recovery from his basic fault. The arglos state, which Balint
(1968) considers to be an absolutely necessary precondition for the

new beginning, is explained by the craving of the patient for primary
love (Balint, 1953). It is clear that the special atmosphere provided
during this state has more to do with recognition than massive

gratification. Only token satisfaction of need is provided, and in the
evolution of Balints views the tokens of direct gratification become
less and less; the recognition of the patients need and the
unobtrusiveness of the therapist are the essential ingredients.
Balints approach to the arglos state is given indirect
metapsychological support by Zavitzianos (1974), in my opinion,
although he does not mention Balint in his paper. Zavitzianos
postulates that an inborn developmental drive of the ego exists, which
propels further development under favorable conditions. When this
ego developmental drive is dominant in the therapy, the ego is not

inclined toward a demand for transference gratification, but rather
toward experience of the understanding and human decency of the
therapist. At such points the offer of direct transference gratification

would be refused and would constitute a complete failure of empathy
on the part of the therapist. For this reason seemingly unanalyzable
patients, explains Zavitzianos, respond to psychoanalysis because the

analyst by his personality and the atmosphere he provides meets the
needs of this ego striving and thus allows further development to

occur from within. This explains in Balints terms why the recognition
and understanding by the therapist of the patients profound needs
and problems produces a new beginning and why attempts to

directly gratify in the transference are actually destructive and miss
the point.
Balint has called attention to the fact that an important

theoretical change in the classical conception of psychotherapeutic
technique is at hand and that this is determined by the increasing
importance given to the actual experience of after-education that the

patient in a benign regression has with the therapist. Such an
experience, of course, is far more important when one is dealing with
borderline or narcissistic personality disorders than when one is
dealing with classical neuroses. Balint recognizes this quite clearly in

his division of types of treatment into those suitable for patients who
are essentially at the Oedipal level and those suitable for patients who
are at the level of the basic fault.
Unfortunately, his theoretical formulation is based on

adultomorphic errors and there is also a mystical aspect to it in terms
of the kind of atmosphere that the therapist is supposed to provide.
All patients should be presented with the physicianly vocation (Stone

1961; Chessick 1977a) and the authentic self of the therapist, and it is
not clear what special techniques are really involved in somehow
trying to provide the patient with totally empathic mothering. We try
to provide all our patients with as much empathically based

understanding as we can, but we are always bound to make some
empathic errors.
It is perhaps more realistic and practical to turn to Modells
technique of allowing a transitional object transference to take place

so that the development of the patient can resume. Winnicott (1965,
1968) introduced the notion of the true and the false self, and this is
associated with Modells ideas. The false self develops in response to

early nonempathic mothering and has to do with learning to be
compliant, a certain inherent rigidity and lack of autonomy or
spontaneous feeling. This often has to be broken down via therapeutic
regression (Chessick 1974a) so that the pathological false self-

compliance can disappear and a real exchange of affect and feeling
can emerge in the therapeutic situation. If the therapist himself is a
true self this will become clear to the patient, however, without any
special alterations in the analytic process.
Kohuts (1971) therapeutic preoccupation involving
regression and transference formation is: How is one to differentiate
the psychopathology of the analyzable narcissistic personality
disturbances from the psychoses and borderline states? For Kohut
the answer seems to rest almost entirely on the type of transference

that is formed when the patient is taken into a formal analysis. Thus, a
differential diagnosis on the basis of initial interviews or
symptomatology is almost impossible, if I understand Kohut

correctly. He writes, . . the spontaneous establishment of one of the
stable narcissistic transferences is the best and most reliable
diagnostic sign which differentiates these patients from psychotic or

borderline cases on the one hand, and from ordinary transference
neuroses, on the other. The evaluation of a trial analysis is, in other

words, of greater diagnostic and prognostic value than are
conclusions derived from scrutiny of behavioral manifestations and
symptoms.
Thus, as previously explained, Kohut places the unanalvzable
psychoses or borderline states on the one hand and the analvzable
cases of narcissistic personality disturbances on the other. The former

tend toward the chronic abandonment of narcissistic configurations
and toward their replacement by delusions; the latter show only
minor and temporary oscillations, usually toward partial

fragmentation.
The schizoid patient, whom Kohut includes among the
borderline cases, keeps his involvement with others at a minimum as
the outgrowth of a correct assessment of his regression propensity
and narcissistic vulnerability. Such patients correctly evaluate their
assets and weaknesses. The therapist should thus not be a bull in the
china shop of the delicate psychic balance of a valuable, and perhaps
creative individual, but should focus his attention on the
imperfections in the defense structures. To put it another way, the
appropriate therapy for schizoid or borderline patients is not formal
psychoanalysis, because a transference regression will take place that
will lead to a severe fragmentation of the self. Instead, a

psychoanalytically sophisticated form of insight therapy is called for
that does not require the therapeutic mobilization of the self-
fragmenting regression.
In this manner Kohut distinguishes among three groups of
patients:
1. The ordinary psychoanalytic treatment patient who forms a

transference neurosis
2. The borderline or schizoid or schizophrenic patient who is an

unsuitable candidate for psychoanalysis and for whom
a regression will lead to self-fragmentation
3.The patient with a narcissistic personality disorder, who forms

certain definite types of stable transference in a
formal psychoanalysis and, thus, is analyzable
Kohut has described what I would call one type of borderline

patient, the type who develops an analyzable narcissistic transference
in a classical psychoanalysis. It does not follow from this that all other
borderline patients are to be labeled borderline schizophrenic or

schizoid patients. To put it another way, it is not reasonable to say
that if a borderline patient is put into psychoanalysis either he will
develop a narcissistic transference of the type described by Kohut or
he will develop a fragmenting regression. Other kinds of transference
can also develop. Whether these other kinds of transference are
amenable to a classical psychoanalysis remains an open question, but

I think it is simply not true that borderline patients who are put on
the couch either develop a fragmenting regression or develop the
kind of narcissistic transference described by Kohut. There are other

alternatives.
This in no way contradicts the formulations of Kohut, with
which I am substantially in agreement. It merely indicates that
borderline patients have various predominating features in their

psychic organization which sometimes show themselves in the
formation of a classic narcissistic transference as described by Kohut,
sometimes show themselves as a regressive fragmentation, and

sometimes show themselves in terms of other kinds of transferences
that may or may not be amenable to analytic interpretation.
Following Kohuts formulations to their conclusions with Gedo

and Goldberg (1973), we can establish a hierarchy of treatment
modalities. Phase One, from zero to six months of age, represents the
time from birth to cognitive self-object differentiation. Primary

narcissism reigns supreme. Primary repression is the crucial
mechanism of defense, and the primary anxiety is that of annihilation
through overstimulation. Patients who have to regress to Phase One
experience traumatic states or panics. The treatment of these cases is

pacification, which represents the control of excitation, controlled
catharsis and, if necessary, the use of medications and hospitalization.
The essence of pacification is tension reduction and mastery through
partial discharge.
A second phase of life, between eight months and three years,
is the phase during which self-object differentiation progresses to

essentially irreversible cohesion of the self. During this phase the
grandiose self and idealized parent imago are utilized, separation
anxiety is the characteristic anxiety, magic is the kind of reality

testing used, and massive projection and introjection are employed.
Patients who regress to such a phase present clinically what we call
psychotic disintegration, and the treatment is that of unification. Such

patients require a cohesion of the self through the therapists
providing reliable and consistently available objects and settings. An
uninterrupted relationship with the therapist is crucial. As the
therapist becomes a transitional object and puts himself in the life of

the patient, there occurs what Balint has called repair of a basic fault,
and the therapy is a real experience for the patient in which he is

having an uninterrupted relationship with a real object. The therapist
sometimes must even force himself into the life of the patient as a real
object.
The third phase of life, from around three to six years, is from
the time of the cohesive self to the solid formation of the superego.
Narcissism becomes more confined to the phallus and castration

anxiety is typical. Disavowal or splitting of the self is the mechanism
of defense, but the self and object are perceived as whole and
different. Patients are characterized by narcissistic personality
disorders when they have regressed to or are fixed in this phase of
life. The treatment then is optimal disillusionconfrontation with
realityand perhaps Kohuts kind of psychoanalysis in which stable
narcissistic transferences are allowed to form and are gradually
interpreted. The patient is helped gradually to give up the narcissistic
fantasies, and the grandiose self and the idealized parent imago are
integrated into the personality.
The final phase of childhood, between six and eight years of

age and puberty, is the phase of consolidation of the ego and the
repression barrier after the superego has been formed. The reality
principle becomes prominent; the person is guided by his ego ideal

and pushed by his ambitions. Moral anxiety is typical, repression
occurs, and we have the era of the infantile neuroses. The treatment
of this kind of disorder is the psychoanalytic method, using
interpretation in which there is strengthening of the ego and

mitigation of the severity of the superego and in which small
quantities of dammed up inner energies are discharged.
One might add a fifth phase of life from puberty to adulthood

that we could call the era of the fully differentiated psychic apparatus.
Signal anxiety is typical at this time, and narcissism has been

transformed to wisdom, empathy, humor and creativity. Difficulties
during this time are hopefully resolved by careful introspection and
even self-analysis (Gedo and Goldberg 1973).
This neat division of the kinds of therapy necessary into

phases of regression or developmental fixation that are appropriate is
not quite satisfactory, because most patients present a mixed clinical
picture. Obviously, patients who present traumatic or panic states
must be given pacification. However, for those patients in the second,
third and fourth phases, the treatment should present interpretation,
optimal disillusion and unification together in the psychotherapy,

with shifting emphasis during the treatment depending on where the
state of regression is in the patient.
The therapist must empathically be able to tune in to where

the patient is and provide the kind of treatment modality that is
optimal for the patient at any given time. The danger of hierarchical
formulations lies in the tendency to fit the patient to the treatment
rather than the treatment to the patient. That is to say, they do not
take into account fluctuations of the ego along the ego axis on a day-
to-day basis (Chessick 1973) in every ongoing intensive
psychotherapy.
Attacks on psychoanalytic psychotherapy that minimize the
importance of interpretation make the same mistake; they tend to

ignore the fluctuations of the patients ego state on a day-to-day basis.
There are times or phases in the psychotherapy where unification and

optimal disillusion, usually loosely referred to as education, are
predominantly necessary. Education is always going on in every
psychotherapy, since there is always, if the therapist behaves
himself (Winnicott 1958), an uninterrupted relationship. On the

other hand, interpretation is also always going on at one level or
another. If it is skillful, it has a more or less important effect
depending on the particular regressive phase that the patient is in at a
given time during the treatment.
This explains the kind of criticism that is constantly aimed at
psychoanalytic psychotherapy by such workers as Strupp. In a recent
paper Strupp (1975) utilizes a case of an at best borderline, probably

ambulatory schizophrenic, patient reported by Balint, and he
demonstrates that interpretation, although claimed by Balint to be the

therapeutic agent, was not the crucial therapeutic agent in the case.
For such patients, as Strupp points out, it is gratuitous to argue that
interpretation is the crucial factor: . . every patient is being influenced

by the therapist in a wide variety of ways, of which interpretations
are only one subject, and ... it is arbitrary to elevate the latter to a
position of preeminence. On the contrary, I believe that the weight of

the therapeutic influence is brought to bear in numerous modalities
and that interpretations of all kinds are a relatively minor factor in
the total change that is wrought over the short as well as the long
term. This kind of argument is used to attack the psychoanalytic
method and interpretation in all cases, and it is based on a confusion

between patients who do not have a cohesive sense of self and
patients who do.
Anyone who works in the area of narcissistic personality
disorders or borderline states must maintain a continual special

awareness of the kind of atmosphere he provides for his patients, and
this is consistent with discussion of optimal disillusion and unification
along the lines of Kohuts theory. Whether this education kind of

treatment is to be labelled psychoanalytic or not seems to me a
semantic question and one which is fraught with overtones of
prestige and status; certainly, no one could deny that it constitutes

optimal psychoanalytically informed psychotherapy.
Part V
PSYCHOTHERAPYSPECIAL
PROBLEMS
Chapter 17
Transference in the Borderline Patient
I. Symbiosis
Much debate about the treatment (and even identification) of
the borderline patient revolves around consideration of the kinds of
transference they develop. These cannot be described in a simple
manner, and furthermore it is not possible to understand the kinds of

transference that borderline patients develop without as thorough as
possible an understanding of the symbiotic phase of development and
the separation-individuation phase, as already described.
Some confusion exists about the exact times of the symbiotic

and early separation-individuation phases. Revision of Mahlers
thinking took place between the 1950s and 1960s; the beginning
emergence from the symbiotic phase was finally believed to occur
earlier. For our purposes we could say that the height of the symbiotic
phase is reached at about six months of age, and from that point on
there begins the separation-individuation; with Mahler we could say
that it is the general task of about the second year of life to reach at
least some solid separation-individuation, although the process is far

from completed and it takes another year yet before there is sharp
differentiation and a sense of self is accomplished. That is to say, only
by the time the child is three years old can we say that separation-
individuation has been passed through in solid fashion.
Now the core of ego development, the first orientation toward
external reality (as Schilder had pointed out already in 1938), is the
differentiation of the body image, which is the psychic representation

of the body self. Through the rhythmically recurring experience of
painful accumulation of tension in the inside of his own body,
followed by regularly repeated experiences of gratification, which the

infant cannot provide for himself hallucinatorily beyond a certain
point, the infant becomes eventually dimly aware of the fact that
satisfaction is dependent on a source outside of his bodily self. So

Mahler (1952) points out that the infant recognizes an orbit beyond
the boundaries of the self, that of external reality represented by the
mother. Bodily contact with the mother, fondling and cuddling, is an
integral prerequisite for the demarcation of the body ego from the
nonself within the stage of somatopsychic symbiosis of the mother-
infant dual unity.
It is very important to try to think about what goes on as the

infant begins to experiment with the feel of his mothers body,
comparing it with the feel of his own, learning about his body
contours as separate from his mothers, distinguishing between
himself and his mother. During the symbiotic stage the mental
representation of the mother remains fused with the mental

representation of the self and it participates in the delusion of
omnipotence of the child.
In the symbiotic child psychoses, as described by Mahler
(1952), unneutralized libidinal and aggressive forces have remained
narcissistically vested in fused systems of mother-father-child unit,
reminiscent of the primary unit (mother-infant). The world is seen as
hostile and threatening because it has to be met as a separate being.

Thus separation anxiety becomes the crucial issue in the symbiotic
psychoses as described by Mahler. The boundaries of the self and the
nonself are blurred.
We get into difficulty when we try to understand what has
gone on intrapsychically during the transition from the symbiotic
stage to the early stages of separation-individuation. Mahler (1975)
feels that during the second year of life the maturational growth of
locomotion really precipitates the separation-individuation. But what
goes on within and among the psychic representations as we shift
from a totally fused self-object representation to separation-

individuation? It is not possible to reasonably assume that there is no
autistic phase and that the child comes at birth with a ready-made
ego, as Fairbairn assumes, or with ready-made capacities for forming
introjects and so on, as Klein assumes, and it seems to me that it is

also stretching things a great deal to depend heavily on the clinical
material of adult patients to make assumptions about self-and object
representations being split in a preverbal child, an infant around one

year of age.
What happens has more to do with a splitting of affects; the
self-and object representations that appear during the therapeutic
process in the transference are already a later accretion to which
these affects are attached. This is more consistent with the cognitive
capacities of the one-year-old than the assumption that self-and
object representations are sitting in his undifferentiated psyche that
can be sharply projected out onto a therapist years later.
Giovacchini (and also Kohut implicitly) differentiates between

adaptative techniques which have been introjected (or as Kohut
would call it, microinternalized), have become part of the childs ego
and do not form a discrete entity like a foreign body, on the one hand,
and disruptive introjects on the other. That is to say, during the phase
of separation-individuation, the ego begins to differentiate itself from
maternal introjects. The child sees himself as separate and distinct.

He learns the mothers adaptative techniques, and this enables him to
achieve further separation and strengthens the ego boundaries. As
this occurs, the maternal introject, if you want to call it that, becomes
part of the childs ego, and insofar as it promotes psychic harmony

rather than disruption it loses its boundaries and becomes
assimilated. This is a functional object relationship; Kohut uses the

complex term transmuting microinternalizations to distinguish these
experiences from the abruptly precipitated and disruptive introjects
which occur when the experiences with the mother are basically
ungratifying.
What happens when these experiences with the mother are
ungratifying during the stage of separation-individuation is first of all

that this stage is not successfully traversed, leaving the individual
with immense separation anxiety, annihilation anxiety and a tendency

to invest massive amounts of anxiety in all kinds of situations
typical of the borderline patient. It also leads to an ego which is
poorly adaptive and has to use a variety of clumsy techniques to deal
with situations that more fortunate individuals are able to handle in a
much smoother fashion.
However, the worst consequence of the ungratifying

experiences has to do with the incredible hatred that is set up, the
relentless, boiling, chronic rage. I see no reason to postulate a
constitutional aggressive factor! Sometimes this rage has to do with

the infants perception of the mothers rejection and hatred of it, but
there are also situations where the mother doesnt hate the infant but
is emotionally absent for other reasons or shifts back and forth or

cannot empathize with the infant because it is needy, dirty, noisy and
so on. The rage and the hatred are absorbed into the personality
structure along with feelings of worthlessness and inadequacy, a
sense of feeling unlovable and vulnerable, a profound lack of self-

esteem and a sensation of being in danger. This sense of danger,
which really has to do with fear of the explosion of hatred into the
conscious mind, is either sensed as danger from some foreign
attacking power within or projected out onto the therapist or others
and experienced as coming from without. Often the wish is to kill this
power, to wipe it out by the use of chemicals or alcohol or to
identify with it and destroy somebody else.
In this situation there are no successful transmuting

microinternalizations, but rather there are what Giovacchini (1975a)
calls hateful maternal introjects, which have a disruptive influence

and which must be kept under control or denied in various ways.
During the process of psychotherapy of the borderline patient, these
hateful maternal introjects, which primarily produce a sense of
combined profound hatred and terrible intense helplessness and
anxiety, may in their various aspects be projected into the therapeutic
situation.
For example, Giovacchini (1972) writes, The patient or child
must maintain a facade of control and autonomy within the
framework of a helpless and vulnerable ego. In order to maintain this

control the child has to be isolated from the threatening introject
which he can achieve by defensive splitting. On the other hand, he
must cling to an external object because of the intense helplessness he

feels. He requires both nurture and rescue from inner assault by the
frightening, disruptive introject. But insofar as his self-representation
includes derivatives of the primitive symbiosis, an inner assault also
seems to emanate from those hateful aspects of the self that are
precipitates of the mother-child fusion. The child then turns to the
outside world for anaclitic nurture and salvation from a raging, self-
destructive self.
This is the foundation of the kinds of transference one sees in
the borderline patient, a turning to the outside world for anaclitic

nurture and salvation from a raging self-destructive aspect of the self.
It is important to keep in mind what we mean by the term introject, a
badly misused term. It is best to think of it basically in terms of

feelings rather than of some kind of personified image or phantasm in
the mind. In the psychotherapy of the borderline patient, what is most

impressive is the unneutralized feelings of all kinds that emerge,
which frighten the patient and which the patient finds very difficult to
deal with. Many such patients dont have these feelings attached to
specific fantasies or these fantasies may change and shift in fleeting
ways in dreams and projections and so on. For example, one patient
would hear or remember the voice of her mother calling her name
in a disgusted tone.
Giovacchini explains that as the patient projects these terrific
feelings onto the therapist, the fact that the therapist responds to the
patient, not with terror or helplessness, but with analytic calm and
interpretations brings the therapists secondary organization to the
patients primary-process chaos. This helps the patient achieve

organization both through self-understanding and by incorporating
the therapist during the regression. Giovacchini (1972) writes, The
regression to the symbiotic phase during analysis can lead the patient
to regain parts of the self that had been split off, and the catalytic
effect of the analytic introject causes them to be synthesized into
various adaptive ego systemsnot to be dissociated as they were in

childhood.
II. Transference
With these considerations of the symbiotic phase in mind, we
can look a little more at the kinds of borderline transference
specifically described in the literature. One aspect of the transference

is the intensity and seriousness of it, as described by Little (1966), for
instance. She begins by agreeing with my basic contention that
borderline state is an imprecise and descriptive term for a wide range

of patients. She emphasizes the sliding back and forth on the ego axis
as I have described it. The separation anxiety in such patients has to
do with fear of annihilation; it is also often accompanied by a literal
psychosomatic chaos. The patient may develop a whole variety of
severe psychosomatic difficulties at the point where he begins, in the
depth of the transference, to experience this annihilation anxiety
(Chessick 1972b)that is to say, the differentiation between psyche
and soma begins to break down and the primitivity of the kind of
anxiety involved makes itself known.
Similarly, as Little points out, the more primitive the form of
anxiety, the more primitive is the form of defense against it and of

course the less effective too, so we see a lot of magical thinking,
omnipotence and simple rigid denial. These are the primitive
defenses against annihilation anxiety: magical thinking, hallucinatory
omnipotence or narcissistic omnipotence and simple denial.
Little stresses the tremendous degree of sensitivity, stability

and flexibility' necessary in the therapist working with borderline
patients, because of the tremendous anxieties that are involved:
Freedom of imagination, ability to allow a free flow of emotions in
oneself, flexibility of ego boundaries, and willingness to consider the

views and theories of colleagues whose approach may be different
from ones own (which is perhaps the same thing) may all prove to be
vitally important in the treatment of any patient. I have tried to

illustrate in the case reports how sometimes the views of one
colleague appear to be most appropriate to a borderline patient; at

other times one applies the views of another. I dont think this is an
accident. It has to do with the variety of patients that are loosely
labeled borderline.
So one aspect of the transference in borderline patients is the
intensity of the annihilation anxiety and the primitive defenses of
magical thinking, denial and narcissistic omnipotence that are used

against it. The other aspect of the transference, which is fairly
consistent, as already discussed from Modell, is the transitional-object
nature of the transferences. The therapist is experienced as an object,

perceived somewhat outside of the self, whose qualities are distorted
by fantasies arising from the subject, and this object, the therapist,
has a real role in the life of the patient in what I would call primarily a
soothing kind of function. Borderline patients will put the therapist in
this role no matter what he says or does, and whether he likes it or
not.
To give a clinical example, a patient comes in session after
session and reels off a long list of irritations and complaints that she
has accumulated during the week. We discuss these various
complaints, sometimes we discuss how she will deal with them, and
she feels better when she leaves. The next session the same thing
happens again. This goes on in an apparently endless series. For a

long time I felt a growing sense of confusion with this patient because
I could not understand why this was happening. The patient simply
didnt seem to be catching on to what psychotherapy was all about. In

fact, she rarely seemed to be listening to anything I said, and yet she
came regularly; she felt that the therapy was helping and her life was
even improving. Finally, after very careful listening, it occurred to me
that what we were dealing with was the equivalent state of a six-
month-to one-year-old infant. A cycle is reenacted in which the infant
awakes from sleep, gradually gets tired, gradually gets hungry,
gradually gets wet, becomes irritable, picks up a variety of complaints
along the way and finally starts crying. Then the mother appears and
cleans it up, holds it, cuddles it and feeds it. Then all is well and the
infant goes back to sleep, and then the cycle begins all over again. The
patient was using me in this sense from week to week as a

transitional object for soothing and tranquillization, touching base to
feel that all was well and I hadnt disappeared or lost interest. Then
she was able to go out and face the outside world, which was
perceived as dangerous, attacking, threatening and separate from the
dyad that the patient had formed of herself and me, experienced as a
part of herself, like a blanket or teddy bear.
If the therapist doesnt understand and catch on to this aspect

of the transference, the result can be a destructive
countertransference, because it is very irritating to be used this way,

as a self-object (Kohut 1971). In the first place, one has the feeling
that one does not have a self of ones own. The patient is not
responding with affectual contact to ones own self but is simply,

deliberately, and selectively ignoring the human aspect of the
therapist and using him as an object. This always stirs up hostility in
another person. In the second place, it all seems to go against what we
try to do in psychotherapy, where we look for the effect of our

interpretations in improving the persons life, leading to better
adaptation, enhanced ego strength and so on. When the kind of
transference I am describing is in effect, however, it doesnt matter

very much what we say! The patient is not interested in the words at
all, any more than when the mother picks up the baby, the baby cares
which lullaby the mother is singing. This is a narcissistic blow to the

therapist, who likes to think of himself as doing something to help the
patient grow and as having thoughts and ideas which are useful to the
patient and helpful in the psychotherapy. This kind of patient is

simply not ready to utilize interpretation.
The therapist has to be aware of the tendency to retaliate and
drive such patients away. If he does not retaliate, transferences which
are characterized by massive annihilation anxiety, by the projection

of tremendous rage and affect onto the therapist, or by the use of the
therapist as an object occur and are often quite stable transferences
when they appear in psychotherapy. Are they workable and do they

respond to interpretation? There is a great deal of debate and
disagreement on this point. Clearly, a tremendous amount of patience
is necessary to work with these kinds of patients. A calm, consistent

approach, not getting sucked into the dramatics, and a consistent
interpretive approach are mandatory. Always staying with the
material, not getting too deep or too fancy or too caught up in
Kleinian terminology over a very long period of time, not only

provides pacification and unification, but also does eventually provide
insight which the patient gradually begins to use in many cases.
It is not possible to predict which patients can get more out of
the psychotherapy than simple pacification and unification, but surely
every patient should be given a chance. Neither the impatience of the

therapist nor the narcissistic blow of being utilized as an object nor
the great slowness of the treatment is sufficient cause to give up on
the patient or to come out with a prescription pad and tranquillize
him.
In my experience from supervision of residents and even with
presenting cases to colleagues, I find two great pitfalls in such

transferences. First of all we must face the therapists fear of these
transferences. It looks as if the patient is exploding, and unless the
therapist has a fairly thorough dynamic grasp of what is going on, he

can become panicky and can be stampeded into doing something
radical or into getting rid of the patient one way or the other. The
second kind of pitfall is impatience. The therapist must be willing to

sit for years with a borderline patient while he gradually catalyzes the
rebuilding of the ego structure. Many therapists simply dont want to
do this, and if they dont, it is probably unwise for them to attempt the
psychotherapy of borderline patients. Every skilled therapist knows

what kinds of patients he works well with and what kinds of patients
he would prefer to stay away from.
Chapter 18
Transference in the Narcissistic Personality Disorder
From Kohuts theoretical structure it follows that the

transferences that arise from the formal psychoanalysis of the
narcissistic personality disturbance will come from the mobilization

of the idealized parent imagothe idealizing transferenceand
from the mobilization of the grandiose selfthe mirror
transference. This depends, of course, on the appropriately attentive
but unobtrusive and noninterfering behavior of the analyst that
Kohut calls the analysts analytic attitude.
In the borderline patient and psychotic, the danger of

regression to the stage of the fragmented self corresponding to the
stage of autoerotism makes mandatory the maintenance of a realistic,

friendly relationship with the therapist and the provision of
psychotherapeutic support, since a workable transference for a
psychoanalysis cannot take place. But for the narcissistic personality

disorder, Kohut (1971) writes:
To assign to the patients nonspecific,

nontransference rapport with the analyst a position
of primary significance in the analysis of these forms
of psychopathology would, thus, in my opinion, be
erroneous. Such an error would rest on an
insufficient appreciation of the metapsychologically
definable difference between unanalyzable disorders
(psychoses and borderline states) and analyzable
forms of psychopathology (transference neuroses
and narcissistic personality disorders).
In the working through of the idealizing transference,

regressive swings take place after each inevitable disappointment in
the idealized analyst, but the patient returns to the basic idealizing
transference with the aid of appropriate interpretation, providing

these interpretations are not given mechanically, but with correct
empathy for the analysands feelings. This leads to the emergence of
meaningful memories that concern the dynamic prototypes of the

present experience. This is the essential paradigm of the working-
through process in the narcissistic personality with the patient that
forms an idealizing transference. It seems clear-cut and clinically
useful.
The therapeutic activation of the grandiose self occurs in the
appearance of the mirror transference, which is more complicated
because it is divided into several types. In its most archaic form the
analyst is experienced as an extension of the grandiose self. In a less
archaic form there is an alter-ego twinship transference in which the
analyst is experienced as being very similar to the patient. In the most

mature and more common form, the analyst is experienced as a
separate person, important to the patient but accepted by him only in
the framework of grandiose needs.
Thus, the mirror transference is the therapeutic
reinstatement of the normal phase of the development of the
grandiose self in which the gleam in the mothers eye, which mirrors
the childs exhibitionistic display, and other forms of maternal
participation in the response to the childs narcissistic exhibitionistic
enjoyment confirm the childs self-esteem and, by a gradually
increasing selectivity of these responses, begins to channel it into

realistic directions.
Which type of mirror transference appears is not as important
as the establishment of a relatively stable transference by the

activation of the grandiose self, for this enables the patient to
mobilize and maintain a working-through process in which the
analyst serves as a therapeutic buffer and enhances the gradual
harnessing of ego-alien narcissistic fantasies and impulses.
The therapeutic mobilization of the grandiose self may arise

either (1) directlya primary mirror transference, (2) as a
temporary retreat from an idealizing transferencereactive
remobilization of the grandiose self, or (3) in a transference repetition
of a specific genetic sequence that Kohut calls a secondary mirror

transference. The regressive swings in the working through are
desirable and cannot be avoided, since no analysts empathy can be

perfect, any more than a mothers empathy with the needs of her
child could be. The understanding gained from therapeutic scrutiny of
these swings is of great value to the patient.
In the treatment of these patients Kohut faces directly the
technical problem of the extent to which the analyst must become
active. He feels that major forceful interference is necessary mainly

in instances of borderline psychoses and in related instances of
profound ego defect that result in unbridled impulsivity. The major
approach to such disturbances or acting out is to alert the patients
ego that a change of behavior is indicated in the interest of self-
preservation. No moral issue must be raised except that practically

and realistically, in view of the prevailing mores, the patient is putting
himself in jeopardy by his doings.
This leads to a crucial discussion of the so-called passivity of

the psychoanalyst during the psychoanalytic treatment, which Kohut
correctly observes has at times been mistakenly discussed as if it
were a moral issue. The essential factors of the process in the
psychoanalytic cure are outlined. A contrast is drawn between

inspirational therapy and psychoanalysis: The former works through
the active establishment of object relations and massive
identifications. Psychoanalytic psychotherapy works through the

spontaneous establishment of transferences and minute processes of
transmuting reinternalization. Kohut explains, If the analyst
assumes actively the role of prophet, saviour and redeemer he
actively encourages conflict solution by gross identification, but

stands in the way of the patients gradual integration of his own
psychological structures and of the gradual building up of the new
ones. In metapsychological terms the active assumption of a

leadership role by the therapist leads either to the establishment of a
relationship to an archaic (prestructural), narcissistically cathected
object (the maintenance of the patients improvement depends
thereafter on the real or fantasied maintenance of this object

relationship) or to massive identifications which are added to the
existing psychological structures.
The formal psychoanalytic process attempts to keep the
infantile need activated while simultaneously cutting off all roads
except the one toward maturation and reality. Only one way remains
open to the infantile drive, wish, or need: its increasing integration
into the mature and reality-adapted sectors and segments of the

psyche, through the accretion of specific new psychological structures
which master the drive, leads to its controlled use, or transforms it
into a variety of mature and realistic thought and action patterns.
Kohut claims that not to make any active moves to foster the
development of a realistic therapeutic bond may be the decisive
factor on the road to therapeutic success; the endless ability to

remain noninterfering while a narcissistic transference establishes
itself is crucial. Furthermore, the manifestations of the inability of
such patients to form a realistic bond with the analyst must not be
treated by the analyst through active interventions designed to

establish an alliance! These manifestations also must be examined
dispassionately.
Turning directly to the theories of Balint, Kohut believes that

imputing to the very small child the capacity for even rudimentary
forms of object love rests on retrospective falsifications and on
adultomorphic errors in empathy. More specifically, he argues that

situations in which the analyst feels that he must step beyond the
basic interpreting attitude and become the patients leader, teacher
and guide are most likely to occur when the psychopathology under
scrutiny is not understood metapsychologically. Since under these
circumstances the analyst has to tolerate his therapeutic impotence
and lack of success, he can hardly be blamed when he abandons the
ineffective analytic armamentarium and turns to suggestion (offering
himself to the patient as a model or an object to identify with, for
example) in order to achieve therapeutic changes.
The calm, well-trained craftsman is held up as the ideal so that,
As our knowledge about the narcissistic disorders increases, the
formerly so personally demanding treatment procedures will

gradually become the skilled work of the insightful and
understanding analysts who do not employ any special charisma of
their personalities but restrict themselves to the use of the only tools
that provide rational success: interpretations and reconstructions.
You can see why it is so important in the theories of Kohut to

make a differentiation between borderline patients and narcissistic
personality disorders. According to this theory, the essence of the
therapy is the formation of a stable transference and the essence of

the cure has to do with the calm, well-trained craftsmans
interpretations and reconstructions. This requires from the patient a
certain cohesive, stable self. Otherwise the therapy becomes
completely immersed in just trying to pull the patient together in

some kind of cohesiveness, and there is no atmosphere in which
interpretation and reconstruction can be made.
However, it seems to me that many borderline patients
provide a sufficiently cohesive self that a similar approach can be

takenthat of a calm, well-trained craftsman based on what
Giovacchini and others call the analysts analytic attitudea cool
objective rational approach to the patients material, without getting
sucked into it and without having to save or rescue the patient. The
skilled work of insightful and understanding psychotherapy can take
place in certain borderline patients who have not at the same time
formed these stable narcissistic types of transferences. They would
not be characterized as narcissistic personality disorders, but they
would not be so fragmented that all you could do with them was
pacification and unification.
Psychotherapy of the borderline patient hinges on a
consideration of narcissistic rage. Kohut (1972) explains, I think that
the overcoming of a hypocritical attitude toward narcissism is as

much required today as was the overcoming of sexual hypocrisy a
hundred years ago. That is to say, We should not deny our
ambitions, our wish to dominate, our wish to shine, and our yearning
to merge into omnipotent figures, but we should learn instead to
acknowledge the legitimacy of these narcissistic forces as we have
learned to acknowledge the legitimacy of our object-instinctual
strivings. Thus it must be carefully explained to the patient that

narcissism is not a dirty world.
Narcissistic rage occurs in many forms, and it
characteristically emerges during the intensive psychotherapy of the
borderline patient when the defensive wall of a pseudotranquillity

which has been maintained with the aid of social isolation,
detachment and fantasied superiorityor chemicalsbegins to give
way. This rage must be tolerated and not retaliated against, even by
sarcastic comments or put-downs by the therapist. It is vitally
necessary for the patient to recognize how the rage emerges when his
narcissistic needs are not totally and immediately fulfilled. These
narcissistic needs come from the grandiose self, which expects
absolute control over a narcissistically experienced archaic
environment and insists on boundless exhibitionism as well as the
exercise of total control.
The dangers of fixed paranoid, depressive and psychosomatic

disorders developing in the borderline patient is always present
because the persistence of chronic narcissistic rage particularly tends
to take place. Kohut (1972) explains, Conscious and preconscious

ideation, in particular as it concerns the aims and goals of the
personality, becomes more and more subservient to the pervasive
rage. The ego, furthermore, increasingly surrenders its reasoning
capacity to the task of rationalizing the persisting insistence on the

limitlessness of the power of the grandiose self; it does not
acknowledge the inherent limitations of the power of the self, but

attributes its failures and weaknesses to the malevolence and
corruption of the uncooperative archaic object. The danger of the
insidious development of a paranoid state is thus evident.
In other patients, this chronic narcissistic rage may shift its
focus from the self-object to the self or to the body self. The result in
the first instance is a self-destructive depression; the consequence in

the second instance may be psychosomatic illness (see Chessick
1972a, 1977b, 1977c). It is very important to be aware of the

intensity and the dangers of this narcissistic rage, to watch as it
develops in the patient's treatment, and to deal with it by appropriate
interpretation rather than retaliative put-downs.
I agree with Kernberg (1974b) that it is important to
differentiate cases where narcissistic rage appears as part of the
initial clinical pathological narcissism at later stages of the treatment.

Patients who show an early and open expression of narcissistic rage
in the initial interviews represent a serious risk for the treatment.
Such patients usually do not do well with uncovering psychotherapy,

and supportive help is the treatment of choice, characterized by
confrontations about the narcissistic rage and the consequences of it
as well as of the primitive tendency to deny it or rationalize it, and by

firm limit setting where necessary. The therapist has to provide
structure for the patient in these cases to help protect him against the
consequences of narcissistic rage.
The therapist functions as an accessory ego to the patient in
these cases, helping to protect him against the dangerous narcissistic

rage which could result at worst in psychosomatic breakdown,
depression and suicide and even at best in ruin to the patients
interpersonal relationships. This protection is provided by
confrontation about the intensity of the rage, firm limit setting when
matters are serious and a constant reminding and prodding of the
patient about the dangerous consequences of this rage. At this point
we have stepped away from intensive uncovering psychotherapy and

moved to a very firm and structured supportive treatment which has
a vital and life-saving function and is mandatory in order to help a

patient who is in serious danger.
The therapist should not attempt to undertake the intensive
psychotherapy of a borderline patient or a narcissistic personality
disorder in uncovering fashion unless he is fairly confident that the

narcissistic rage is within bounds that will not result in a catastrophic
destruction to the patient of one form or another. He must be

constantly aware of the dangers of the eruption of such rage, and
when such eruption threatens he must be alert to it and help the
patient to deal with it. That is one of the first priorities in the
intensive psychotherapy of the borderline patient.
Chapter 19
Ambience of the Treatment
Primitive Anxiety.
I wish to turn now to a greater refinement of our concepts,
which will lead to a more sensitive concern with the ambience of the
psychotherapy of the borderline patient. In the situation of the
borderline patient the ambience of the treatment is extremely

important. In fact, it probably constitutes the major nonverbal factor
that determines the success or the failure of the therapy. One could
distinguish between the characteristic affects that appear in the

psychotherapy and the characteristic affects (Mahler and Gosslinger
1955) of the phase of regression or fixation that the patient is in at the
current point. So for example, if the patient has regressed to or is

fixated in the phase of separation-individuation, the characteristic
affect that is experienced by the patient is an intense sadistic rage,

which usually appears as sexual and anal sadism, often combined. For
example, the patient may be much preoccupied with the tearing up of
a person anally, in one way or another, as in Freuds famous case of
the rat man (Freud 1909). The patients preoccupation was with a
story that he heard about a form of torture: a cage of rats was put on a
mans buttocks and the rats burrowed their way or ate into his body.
This enormous and primitive anal-sadistic and sexually sadistic rage

is combined with profound separation anxiety. The kind of anxiety
and upset that appears in these cases hinges on the issue of
separation from the object, upon which the patient is extremely

dependent and attached.
If we move a step backwards developmentally into the
symbiotic phase, the rage tends to take more oral-sadistic form, with
fantasies of cannibalism, biting and tearing and so on. The anxiety
also becomes somewhat different and is expressed as a fear of literal
annihilation, although it may be annihilation through abandonment. It
doesnt have to be annihilation through abandonment; it also could be

annihilation as retaliation for oral-sadistic destructive fantasies. Thus
the patient may fantasy a kind of apocalyptic end of the world in
which he is machine-gunning people right and left, tearing them to

bits, and finally he is himself destroyed. Some psychotic patients have
literally acted this out.
Then as we move towards a phase even farther backwards
(whether we can really reach it or not is certainly moot), there

appears a more autistic kind of situation. The patient displays a kind
of pseudoserenity, which is based primarily on a sense of
hallucinatory omnipotence. If this is disturbed in any way, what
appears is a kind of massive undifferentiated rage discharge, and the

anxiety appears to be of less consequence at this point than the rage,
which, when it appears, is almost totally without overt psychic
content, but rather appears as body expressions and behavior in a
kind of an undifferentiated temper tantrum. Most psychotherapists

do not work with this kind of patient in the office.
It is important to keep in mind the clinical fact that separation
anxiety is not synonymous with the fear of annihilation. Separation
anxiety is somewhat less abruptly overwhelming and is more complex.

By paying careful attention to the kinds of rage and anxiety being
expressed, one has extra diagnostic help as well as an extra indicator

of just where the patient is at any given time in the psychotherapy.
It should be evident by this time that I consider the key to
understanding the borderline patient to be our understanding of the

intrapsychic contents of the symbiotic phase and the early
separation-individuation phase. The aim and successful outcome of
the separation-individuation process is, as is generally agreed, a

stable or cohesive image of the self. Memory deposits within the
inborn and autonomous perceptive faculty of the primitive ego tend
to occur and coagulate into what Mahler calls little islands within the
hitherto oceanic feeling of complete fusion and oneness with the

mother in the infants semiconscious state. These memory islands are
not allocated either to the self or to the nonself. They are primitive
memory deposits of feeling, either pleasurable-good or painful-bad.
Because the experience is repeated in psychotherapy, it is
most important to keep in mind that the young infant is exposed to
rhythmically and consistently repeated experiences of hunger and

other need tensions arising inside the body that cannot be relieved
beyond a certain degree unless relief is supplied from a source
beyond the infants own orbit. This repeated experience of a need-

satisfying good outside source to relieve the infant from
uncomfortable or bad inside tension eventually conveys to the
infant a vague affective discrimination between self and nonself.
Arguments arise at this point; when the discrimination
between self and nonself begins, do good or bad memory islands
become vaguely allocated to self and nonself? Many authors think

they do. The general trend of the authors assumes that confluence
and primitive integration of scattered good and bad memory islands
into two large good and bad images of the self, as well as into split
good and bad part images of the mother, occurs somewhere around
the end of the first year of life. Mahler feels that this is attested to be
the normal emotional ambivalence that is clinically discernable

during the second year of life.
We have, then, rapidly alternating primitive identification

mechanisms, leading to what has been described by Klein and the
modified Kleinian followers as projection and introjection, in which
the infant attempts to deal with these images. As I have repeatedly

pointed out, this appears to be too fanciful to me and assumes a
greater cognitive capacity in the mind of the one-year-old than seems
reasonable. It represents a kind of personification that we as adults
make when we observe the clinical phenomena.
Projection.
In the dialogue of the session, the patient talks of parents,
friends, other people and the therapist without realizing that he is

discussing aspects of himself. Through interpretation we hope that
the patient eventually realizes his identification with the other and its
connection with significant defenses against anxiety. The patient
projects out certain unacceptable aspects of himself onto others
around him, which enables him to become more comfortable. Then
the patient reacts to others as if this projection was of the principle

feature of the other person, utterly ignoring the true personality of
the other individual.
Here is a clinical example: The patient at one point during her

psychotherapy insisted that I was extremely harsh, critical and
intolerant of her; this arose rather suddenly and in the context of a
previously warm and strong therapeutic alliance. A considerable
discussion took place of the patients own harsh, critical and

intolerant superego system, which really represented the internalized
aggression that she had against her extremely disappointing parents.
In the midst of this discussion the patient reported the following

dream: Dr. Chessick was giving a seminar and I was there with
several other people. They became increasingly upset with what he
was saying, and after the seminar we were discussing the content of
his presentation. The others felt that he was very harsh and critical,
but I reassured them he was not really that way at allthat actually
he was warm and understandingand after they get to know him
they will see that they have made a mistake in their judgment of him.
Thereupon they accused me of being very harsh and critical and they
begin to argue with me bitterly.
Now of course, dreams can be interpreted out of context in
many ways, but I have introduced this dream as an example of a
critical working-through dream of a patient who had made a
projection. My having done so presupposes that the therapist in this
case was not in reality harsh and critical. For if he was, then the
patient had a legitimate right to rupture the therapeutic alliance, since
the therapist was showing a lack of empathy with the patients
difficulties.
In the psychotherapy of the borderline patient, one must be
most careful not to fool oneself, when negative images of the therapist
appear in dreams or overt material, into thinking that it is a
projection, if there is solid reason for the patients complaint! Patients
are very intuitive and often present a picture of the therapist that he
does not particularly want to know. Unfortunately, they tend to pick
out the negative aspects of the therapist and emphasize these, while
ignoring the positives. ( I will discuss this in greater detail a little later
when I bring up once more the concept of externalization.)
Adler (1973) utilizes projective identification (see Chapter 9)

and splitting mechanisms to understand the behavior and the
problems that staff members have with hospitalized borderline
patients. He describes an approach to the treatment where the staff
attempts to understand its own retaliatory fury toward these

patients, a fury often aroused by their provocative behavior. The staff
in the hospital has to be helped to set limits in a nonpunitive way, and
the problem is essentially the same as the one that arises in the office
treatment of the borderline, except that it is more acute.
Adler points out that because these patients exhibit so many
areas of strength and even appear sometimes to be psychoneurotic
when they are first seen, behavioral regression, that is to say,
provocative and manipulative behavior, is often interpreted by the
staff as willful misbehavior. The patient is experienced as a bad child
rather than as an overwhelmed patient under great stress. These

patients are expert at devaluating and provoking staff members and
making them feel helpless. The concept of projective identification is
useful to understand the patients attempts to get rid of a part of

himself by placing it into an object, which may then reallyout of
retaliation persecute the patient. Another aspect of projective
identification is that the patient has to extend much effort and activity
to control the person who is the recipient of the projective part of the
patient, for if the patient does not control this person, he then feels in
danger of being overwhelmed by the part projected onto that person.
Adler also notes that on the psychiatric unit, different staff members
may be the recipients of different split parts of the patient. The

patient is actually like a chess player, as I see it; he splits off various
aspects in the hospital and relates them to and projects them onto
various staff members. Then he even manipulates the various staff

members to act in a way which is essentially consistent with the role
they are supposed to play in terms of their being the recipients of

these split parts.
I cannot stress enough how important it is, in dealing with
borderline patients, to be aware of their tendency to set up in external
reality the kinds of situations they need to have occurring. Sometimes
they are quite expert at this, and the therapist almost finds himself
sucked into playing various kinds of roles, depending on the
projection assigned to him. Please notice that the first reaction to this
kind of maneuver is retaliation. Nobody likes to be used as an object,
and nobody likes to be manipulated and forced into a role, especially

a role that is negative and that they dont want to play. Therefore, in
dealing with borderline patients it is almost invariably an error to
take as a personal attack on oneself this kind of behavior; it is far

more therapeutic to try to understand what is going on and to reflect
it back to the patient. The worst possible approach is to become very
defensive when one is accused of all sorts of negatives that are

projected, even if a kernel of truth exists in the accusations. This in
turn requires a thorough self-understanding from the therapist and a

reasonably healthy therapist who is not thrown into an anxiety panic
when his defects are pointed out to him.
The point is not that the therapist should have no defects, the
point is (1) that the therapist should have only a normal amount of
defects, and (2) that he should be reasonably aware of and

comfortable with his defects, so that when the borderline patient
seizes upon these or plays up to them, he doesnt fall into the trap of
losing perspective on his own selfwhich is what the borderline
patient wants him to do. The therapist recognizes that he has some of
these defects but that that's not all there is to him, and therefore he
doesnt feel that he has to make up to the patient or defend himself
against the patient; he is aware that selective perception and

projection are going on.
Obviously, it is impossible to argue a borderline patient out of

the accusations that he makes against you. Only two possible roles are
reasonable. First, if on objective assessment it turns out that the
accusations are correctand this sometimes happensthen the
therapist needs to correct himself. If, as is hopefully the case, the
therapist is reasonably healthy, we are dealing with projection. The
proper approach to this is a calm, nonanxious and patient stand, with
eventual interpretation of what is happening. It is this calm,
nonanxious and patient stance that provides the basic ambience of
the treatment. Any disruption of it interrupts the subliminal soothing

that is always going on in a well-conducted treatment of a borderline
patient. No matter how we wish to get away from this in our
theoretical conceptions, the ambient subliminal soothing the

therapist provides in his habits of consistency, reliability and
integrity; in the ambience of his office; in his personality; in his deep
inner attitude towards his patients, which cannot be fakedprovides

the basic motor that permits the psychotherapy of the borderline
patient to go forward.
Chapter 20
Rage and Externalization
From Kernbergs (1974b) point of view, borderline personality

organizations which do not form the classical narcissistic
transferences described by Kohut may still be amenable in many

instances to a formal psychoanalytic approach with parameters as
the treatment of choice. It should be noted that the Kernberg-Kohut
debate is about the classical psychoanalysis of these patients; the
minute one introduces so-called parameters, the debate shifts. I am at
least fairly certain that Kohut considers himself to be talking only

about classical formal psychoanalysis; Kernberg seems a little more
prone to modifications of psychoanalytic technique when necessary
because he is not concerned with fostering the development of

Kohuts classical narcissistic transferences (Kohut also introduces the
modification of not interpreting the idealization of the therapist, but

this is relatively minor.)
Kernberg and I clearly differ on the optimal treatment for most
borderline patients. He recommends attempting formal
psychoanalysis with as few parameters as possible, whereas I feel

that psychoanalytically informed psychotherapy twice or three times
weekly is most effective. Still, we agree on one major aspect of the
treatment, as he describes it (Kernberg, 1975a): . . . the patient must

come to terms at some point with very real, serious limitations of
what life has given him in early years. ... It is probably as difficult for
borderline patients eventually to come to terms with the fact of
failure in their early life as it is for patients with inborn or early
determined physical defects to acknowledge, mourn, and come to
terms with their defects. Borderline patients gradually have to
become aware of how their parents failed themnot in the distorted,
monstrous ways which existed in their fantasies when beginning

treatment, but failed them in simple human ways of giving and
receiving love, and providing consolation and understanding, and
intuitively lending a helping hand when the baby, or the child, was in
trouble.
It is questionable whether classical transference neuroses
occur in uncovering psychotherapy (as opposed to a formal
psychoanalysis) in a substantial enough form to be amenable to

interpretation. Because of the mandatory increased activity of the
therapist (H. Friedman 1975) especially in uncovering psychotherapy

with borderline patients, other types of transference tend to
predominate in many instances. I am inclined to agree with Kernberg
that, on the one hand, numerous borderline patients placed in the

modified psychoanalytic situation do not regress and fragment into
an open psychotic state; on the other hand, they do not form the
classical narcissistic transferences of Kohut.
This may be a function of the personality of the therapist or of
his entire approach. As described in my books (1969, 1974b) such

patients can form workable transferences which are amenable to
interpretation. Often they cannot, however, and so they can respond
only to some form of supportive psychotherapy which aims at giving
the patient a better structure in dealing with internal and external
adaptations and conflicts.
Clinically speaking, the big problem with many of these

patients boils down to helping them deal with their paranoid feelings
and their tremendous rage. Actually the explosions of rage, as painful
as they are for both patient and therapist, are not as serious to deal
with as calculated (conscious or unconscious) retaliatory attacks over
a long period by the patient on the narcissistic defects in the therapist,
as discussed in the next chapter.
Sometimes the rage of the borderline patient is stirred up
directly by frustrations of his need for omnipotent control of

everything; sometimes the rage is a secondary phenomenon to a
paranoid projection or a transference projection in which, if someone
criticizes the patient or interferes with one of his plans, the instant
reaction is that someone hates the patient. The patient then reacts as
anyone would to someone who hates him. The recipient of the
patients reaction is often surprised and stunned and finally goaded
into retaliatory behavior by the patients clearly hostile and

provocative action.
Thus in a sense these patients are correct when they predict
that all human relationships will end up badly for them, with
disappointment and dislike coming from everyone around them. I

have spoken of this in another context (Chessick 1972b) as
externalization in the borderline patient, a phenomenon in which the
patient responds selectively to the negative aspects of the people
around him and develops a case based on selective negative
perceptions for expecting attack from all sides. The chronic calculated
attacks on the therapists defects, if not interpreted, can lead easily to

countertransference acting out on the part of the therapist, even to
the point of getting rid of the patient. This is quickly worked into
proof by the patient of his expectation of apparently unprovoked

betrayal and abandonment.
Credit for coining the term externalization is usually given to
Anna Freud (1965). She described externalization as a subspecies of
transference and separated it from the transference. Her main

experience with externalization is, of course, in the analysis of
children, and she sees externalization in child analysis as a process in
which the person of the analyst is used to represent one or another

part of the patients personality structure.
The concept of externalization was made a great deal more

precise with respect to the psychotherapy of borderline patients with
deep narcissistic problems in a paper by Brodey (1965). Brodey

points out that in his experience working with family units,
externalization appeared as a mechanism of defense defined by the
following characteristics: (1) Projection is combined with the
manipulation of reality selected for the purpose of verifying the

projection. (2) The reality that cannot be used to verify the projection
is not perceived. (3) Information known by the externalizing person is
not transmitted to others except as it is useful to train or manipulate
them into validating what will then become the realization of the
projection. In other words, externalization makes possible: A way of
life based on relationships with unseparated but distant aspects of the
self. What is perceived as reality is an as-if reality, a projection of
inner expectation. The senses are trained to validate; the intense
searching for what is expected dominates and enforces validation. It
is difficult not to validate an unquestionable conclusion. Each
validation makes the conclusion even less questionable. The

restricted reality perceived is experienced as if it were the total
world.
The psychotherapist feels the intensity of his patients effort to

manipulate him into validating projections. He feels the conflict as he
struggles against this manipulation, but behavior that will be used as
validation seems the only way to gain relationship with the patient.
Thus, the manipulation of the therapist into behavior that is
symmetrical with the projection is different from the simple transfer

of feelings to a therapist.
Even if the therapist does not wish to conform, he still finds
himself conforming to the narcissistic image. For no matter what he
does, pieces of the therapists actual behavior irrelevant to the
therapists self-identity are seized on by the patient, to whom they are

predominant as-if characteristics. The identity that the patient sees
may be unknown to the therapist (although it holds a kernel of truth
which usually is disturbing to the therapist) (Brodey 1965). Even the

therapists active denial of the patients presumption is used by the
patient in the service of proving to himself that the therapist actually
is congruent with his projective image.
Brodey points out that the therapist of the ego-disturbed
patient must become skilled at managing his congruence with the
patients projected image. This management is often intuitive and

usually very demanding emotionally. Being a distorted object is
much easier than being nonexistent.
Fundamentally, externalization is projection combined with
and followed by selective perception and manipulation of other

people for the purpose of verifying the initial projection. Other people
are experienced wholly in terms of their value in verifying the initial
projection, and only those aspects of other people which have this
value are perceived at all. Thus, the most benign therapist
approaching the borderline patient finds himself transformed into a
horrible monster very quickly by the patients selective perception,

and unless he is aware of this danger he is inclined either to retaliate
or to quarrel with the patients extremely unflattering image of him,
which usually contains a kernel of truth and is a direct assault on the
therapists narcissistic conception of himself as a benevolent

physician.
Giovacchini (1967b) emphasizes paradoxical self-defeating

behavior with a defensive purpose, which is usually the result of
externalization. This must be distinguished from self-defeating
behavior resulting from a breakdown of the personality. Patients of

the former type cannot cope with a warm and nonthreatening
environment: They react to a benign situation as if it were beyond
their level of comprehension. These patients do not have the adjustive
techniques to interact with a reasonable environment. Their
formative years were irrational and violent. They internalize this
chaos and their inner excitement clashes with their surroundings.

When the world becomes benign and generous, the patient withdraws
in panic and confusion.
Giovacchini points out that the patient expects and brings

about his failure and adapts himself to life by feeling beaten in an
unpredictable and ungiving world. He distinguishes this from a
masochistic adjustment and points out the relationship of
externalization to the repetition compulsion upon which it is based.
Thus, when the therapist presents the patient with a

consistently benign environment, one which Winnicott (1958) has
described as being parallel to the healthy maternal environment, the
patient cannot trust the lack of frustration. To risking the inevitable
disappointment that he expects, the patient prefers relating in a

setting in which he has learned to adjust. If the analyst does not
frustrate him, the patient's psychic balance is upset. To reinstitute ego

equilibrium the patient attempts to make the analyst representative
of the world that is familiar to him.
Externalization is not simply a projection of internal aspects of

the personality onto the therapist; it contains also a mode of
adaptation or adjustment that makes any other interaction between
ego and the outer world impossible. As Giovacchini (1967) points out,
Externalization provides the patient with a setting that enables him
to use adjustive techniques that he has acquired during his early
development.
Such patients, therefore, are often in a rage for one reason or
another, and techniques must be found to help them calm down, for
such rages directly interfere with their functioning and provoke
retaliation from those around them. When the patient is unable to
gain insight into this crucial problem, the psychotherapist may have
to be satisfied with the development of tranquilizing techniques that
the patient must learn. One way to do this is to help him regard his
rages as a fever that comes upon him, which demands treatment as an
illness rather than viewing his rage as justifiable and appropriate.
In a few cases I have seen, severe obsessive rumination

appears instead of these rages. The net result is the same in that the
patient is functionally paralyzed, although less retaliation is provoked
by the rumination.
In all of these cases, helping the patient in psychotherapy to
become acquainted with his grandiose self, his search for an idealized
parent, his tendency to regard others as self-objects, his paranoid
projections and the continual rages that ensue, and his chronic and
self-damaging narcissistic rage forms the core of the psychotherapy.
As Kernberg (1974a) explains, the problem of rage poses the

greatest danger to the psychotherapy of the borderline patient: The
relentless nature of this rage, however, the depreciatory quality that
seems to contaminate the entire relationship with the therapist, and
what evolves as a complete devaluation and deterioration of all the
potentially good aspects of the relationship for extended periods of

time so that the very continuity of treatment is threatened, are
characteristics of narcissistic patients functioning on a borderline

level." To put it another way, the autonomous ego of the patient must
form an alliance with the therapist if the therapy is to proceed

successfully; the outbursts of such narcissistic rage tend to submerge
the autonomous ego and rupture the therapeutic alliance either
directly or by the production of acting-out or serious symptomatology

or psychosomatic disorders that make treatment impossible in an
analytic setting.
Loewenstein (1972) points out that psychoanalytic treatment
requires the patient to have some degree of integrity of the ego:

This means intactness not alone of some defenses, but also of
autonomous functions," for the autonomous ego is the medium
through which patients communicate to the analyst what they

observe in themselves." The analytic setting, with its frequent
sessions and use of the couch encourages the relative increase of

primary-process thinking and the mechanisms of displacement and
projection, and yet the basic rule also requires the patient to
communicate all his resulting self-observations in a way that is

intelligible, as only secondary process allows it. More precisely, the
autonomous ego must ally itself with the analyst. This requires
relatively intact memory, thinking, perceptions, reality testing,

capacity for self-observation and for verbal expressions.
The autonomous ego must be available for the patient to
become aware through appropriate interpretation how each minor
empathic failure by the therapist or each inevitable frustration in the
psychotherapy (such as the therapists changing an appointment or
taking a vacation) produces narcissistic rage and tends to produce in
addition a psychic switch from searching for the idealized parent-
therapist to withdrawal into the grandiose self (Kohut, 1971). Clinical
manifestations of this switch, along with overt narcissistic rage, are:

coldness toward the therapist, a tendency to primitivization of
thought and speech (from stilted speech to neologisms and
grammatical peculiarities), attitudes of superiority, a tendency

toward increased self-consciousness and shame propensity (due to
increased exhibitionistic tendencies of the grandiose self) and
hypochondriacal preoccupations. It is necessary for the clinician to be

exceptionally alert for the occurrence of these manifestations.
If narcissistic rage shatters the therapeutic alliance, of course
achievement of insight cannot occur. Similarly, if the therapist reacts

to these manifestations with rage of his own, a complete
metapsychological misunderstanding, usually based on
countertransference, has taken place and the therapy is destroyed.
Chapter 21
Countertransference
Clearly, countertransference is of great importance in the

psychotherapy of both the borderline patient and the patient with a
narcissistic personality disorder, regardless of how one

metapsychologically separates these two kinds of patients. The
difficulties in the path of therapeutic dealing with such patients are
enormous. Even Anna Freud (1969) in her famous discussion of
difficulties in the path of psychoanalysis has mentioned the problems
and divided them into difficulties coming from the external world,
difficulties within the patient and difficulties within the analyst. For
instance, the external-world difficulties include the fact that
psychoanalysis has to compete with other therapeutic modalities; the

patient is often under tremendous pressure from friends and relatives
not to come so long and to beware of the therapistthe news media

are filled with discussions of the exploitation of innocent patients by
unethical therapists, and even some young people today tend to see
psychoanalysis as a conservative force, designed to adjust people to

society as it is. There are similar pejorative implications toward
psychoanalytic psychotherapy. To this we have to add the stigma, the
expense, the stress, the time loss and so on, all of which create
external-reality problems for the patient who is involved in any long-
term psychotherapeutic efforts.
The difficulties within the patient are probably even more
important. Anna Freud labels such factors constitutional, involving
adhesiveness or rigiditya weakness of the ego on a constitutional

basis.
Psychological sources working against the cure within the

patient include the need for punishment, intolerance to anxiety, the
innate incapacity to accept substitute gratifications and find suitable
sublimations for the drives and the inevitable inner conflict between
the wish to develop or grow up versus the desire to hold on to the
past.
Although some authors speak of this as the adhesiveness to
the past, we also have to realize that no matter how poor and
inadequate the patients solutions are, they are his production, they
are his creation, they were formed by him as a child during the highly
narcissistic phase, and they are extremely invested with narcissistic
libido. Therefore, there is a certain unconscious pride in these
solutions no matter how poor they may be on a realistic basis, and it
is always a narcissistic blow to the patient to give up long-standing

patterns of adaptation that he has developed under hardships and
with much effort, even though he is giving up these patterns for
something better. There is also a risk involved, because sometimes

the patient has to depend at least to some extent on the faith of the
therapist that the new patterns of adaptation are going to be better.
The patient may know it intellectually, he may see that it works for
those around him, he may hear the faith of the therapist, but he
himself has not experienced it. So there is unquestionably an
unknown-risk factor that tends to cause a conservative tendency in
the patient and a tendency to fall back on tried and tested patterns, no
matter how poor they may be.
In addition to this, the factors within the patient have to do

with the developmental phase during which the crucial difficulties
began. When you are dealing with disasters in the first year or two of
life, you are making an assumption that the damage such archaic
events have caused in a patients psyche can be reversed or at least
ameliorated by psychotherapy. This is as yet an assumption not
generally agreed upon. In addition, the role of constitutional factors

becomes especially crucial in the very early months of life. Rates of
development of the nervous system, constitutional endowment and
so on have a great deal to do with how the very young infant deals
with stimuli. Thus the farther back you go in patients with early
developmental disorders, the more vagueness, confusion, and
disagreement there is about what happened and how to make it

better.
Finally, there are the difficulties in the therapist himself. This
is the source of difficulty for which we have the most direct

responsibility. It is important to realize that we have an ethical
obligation to develop the maximum understanding of the limitations
of our technique and of the countertransferences that arise when we

work with borderline patients, and we also have to realize that the
two are connected. When you work with patients of this type, and you
begin to bump up against these limiting factors, a typical and specific
variety of countertransference is produced.
Grinker (1955) claims, What is not treatable and not
analysable are the effects and results of the first vital non-conditioned
reflexes when mother-child ego-non-ego are not differentiated. This is

the nonreducible residue like the amorphous dust of a ground-up
object of art. ... The effect is in the tissue and analysis cannot modify
it... . What we can do is to . . . loosen, decrease, or modify the learned
function; to get as close as possible to the primary, narcissistic,
depressive, or psychosomatic core; then to help rebuild and

reconstitute more adaptive assemblages of defenses and syntonic
expressions.
In this task, new technical problems confront the therapist.

The closer he approaches to the basic core functions, the more
difficult and dangerous are his problems. Grinker explains, He is
confronted with primary processes and an ego which seems unable to
perform its functions of self-discrimination, reality-testing, or

synthesis. At this point the alert analyst then questions himself as to
whether he is pursuing a harmful procedure which could destroy the

defensive capacity of the ego against the development of a psychosis
or the liberation of serious suicidal trends or the acting-out of asocial
or antisocial behavior. It is at this point that we restudy our material

of the first interviews and our diagnostic workup to determine
whether we have overlooked latent psychosis of some kind that is
now threatening to become active and dangerous. These are the kinds
of anxieties that must be stirred up in the alert and conscientious
therapist even in a well-conducted treatment of the borderline
patient.
Borderline patients and narcissistic patients are especially
vulnerable to acting out manipulations, exploitations, retaliations and
seductions from psychotherapists who are untrained and untreated.

But even well-trained therapists tend to fall into some of these
countertransference problems. For example, Grinker (1955)

mentions, The countertransference may be maternally seductive
towards deep or rapid regression and a high value placed on
therapeutic, hopefully temporary, dependency. Sometimes

countertransference attitudes may consist of firmness and tacit
urging towards growth and change either at the onset or too soon
after regression has developed. Some analysts may vacillate between

these two attitudes as many parents do, seducing dependency and
urging growth at the same time. It is very important for the
psychotherapist to keep in mind, as I have stressed in my books on

psychotherapy (Chessick 1969, 1971c, 1974b), that every patient has
a kind of internal timetable, an internal unfolding program that must

not be hurried regardless of the need of the therapist or the external
pressures on the patient! Attempts to hurry it simply result in failure.
Probably the best analogy is to the timetable of a young
teenager trying to catch on to algebra. The development and the

unfolding of the cognitive apparatus determines the point at which
even the reasonably motivated teenager grasps the concepts of
algebraof substituting letters for numbers. This is consistent with

Piagets discussion (See Evans 1973) of the developmental phases of
cognition and thought. Time is required to work these things through,
and the therapist must not hurry them or push them on the one hand
or hold them back on the other. A well-conducted psychotherapy
implies that the therapist gradually gets the picture of how the
patients internal timetable works and gradually understands the

specific patients techniques of learning, cognition and assimilation,
which differ biologically from individual to individual. This is the
greatest protection against countertransference pushing and shoving

of the patient.
As our knowledge of psychotherapy with the borderline
patient increases, it is apparent that the attitude and reaction of the
therapist are of much greater importance in the treatment situation

than is the case in the treatment of neurotics. Therapists who work
with borderline patients cannot avoid from time to time experiencing
extreme and intense anxiety and suffering the indignities of being

ridiculed, scorned, ignored, disarranged and verbally assaulted.
Physical assault is a very unusual event and has diagnostic
considerations for psychosis involved in it, but verbal assault is

ubiquitous in the treatment of such patients.
Furthermore, borderline patients often note the anxiety of the

therapist and seize on it for externalization. Although we advocate an
objective, analytic and nonanxious approach, we realize that from
time to time such patients are bound to make the therapist anxious.
One of the most important factors in the treatment of the borderline
patient is how the therapist deals with his own anxiety. This forms an
adaptational model that the patient can incorporate, or introject, or
identify with, depending on what terminology you wish to use. It is

unavoidable in psychotherapy, and I think the subliminal observation
that the patient makes about such things as how the therapist deals
with his own anxieties in the treatment has an important effect in
terms of whether the treatment moves forward or not.
You cannot teach this to a therapist; you can only tell him to
get an intensive, thorough psychotherapy of his own, so that his ego

mechanisms and adaptational techniques are as healthy as possible;
then you have a psychic field to offer the patient that is as healthy as
possible. Most authors agree today that the goal of the therapists
personal analysis and training is not to eliminate countertransference
problems, but to shorten the time required for their recognition and
resolution. It is impossible not to develop countertransference

reactions.
It is well known that borderline patients frequently threaten

suicide and even make desperate gestures. This produces a special
kind of countertransference. Maltsberger and Buie (1974) point out,
The countertransference hatred (feelings of malice and aversion)

that suicidal patients arouse in the psychotherapist is a major
obstacle in treatment; its management through full awareness and
self-restraint is essential for successful results. The therapists
repression, turning against himself, reaction formation, projection,

distortion, and denial of countertransference hatred increase the
danger of suicide. I think in all fairness and honesty we have to
report that the psychotherapist who wishes to work over many years
with a large practice of borderline patients has to be prepared for the
eventuality that from time to time, a patient will either directly or

even by accident successfully destroy himself. When this happens it is
always a terrible experience for everybody concerned, and dealing
with this kind of experience is a test of the psychotherapists
personality.
It is remarkable, in my own clinical experience, to observe the
reaction other psychiatrists have when they discover that a

psychiatrists patient has committed suicide. The most striking
reaction I have seen is omnipotent denial: I cannot understand why it
happened; it never happens to me. A therapist dealing with

borderline patients will have such a crisis from time to time, but it
will be a rare occurrence. A therapist may practice for many years
without any borderline patient successfully committing suicide. (If it

appears that a large number of the therapists patients are attempting
or successfully committing suicide, then obviously we have a different
problem.)
When suicide happens, as it does from time to time, it stirs up
tremendous problems, both realistic and internal, for the

psychotherapist. One of these problems is the narcissistic blow to the
therapist. We all know that the normal mourning process includes
with it a rage at the person who has left us. No matter how fine that
person may have been, we have suffered a narcissistic loss and we

rage about it.
Therapists with narcissistic problems tend to ward off anxiety
over the very passivity of the psychotherapy situation and certainly
over the helplessness a therapist feels when his patient has

threatened suicide. They usually tend to ward it off by such behavior
as excessive verbal activity, prescription writing or
overaggressiveness. Murphy (1973) explains how such therapists
may misuse and overcharge patients considerably, and sometimes
they undercharge them for the same reason. Patients find it is easier
to entice narcissistic therapists into playing roles assigned to them via

projective identification and even attempt to seduce such therapists,
who are flattered by the affectionate attention of patients.
It is almost superfluous to add that when a borderline patient

or a patient with a narcissistic personality disorder is being treated by
a psychotherapist who is himself a borderline patient or a patient
with a narcissistic personality disorder, the situation is bound to end
up in something destructive or, at worst, suicide. Most commonly it

ends up with acting out, sometimes individually, sometimes mutually
and together. I have to add this because although it seems obvious, it
is not so rare. I have run into a variety of cases in many years of

practice with borderline patients or patients with narcissistic
personality disorders who have been treated by psychotherapists
(and the variety of these is endless, including social workers,
psychologists, ministers and psychiatrists) who are clearly borderline

patients or are at best patients with narcissistic personality disorders
themselves; there has occurred a mutual seduction, a mutual acting
out, and the events that follow are often cataclysmic and always
destructive in one way or another.
I must sadly record that at the present time there are
absolutely no standards in the United States that license or dont
license a person to practice psychotherapy. Since there are far more
borderline patients desperately in need of help than there are skilled
psychotherapists, the unskilled, the sick, and the acting out, enticed
by money and gratification, are rushing in to breach the gap. It
remains the ethical duty of anyone who practices psychotherapy to
press for social legislation to deal with the training and licensing of
psychotherapists. That is the very least we can do about this present
and dangerous problem, which at times is absolutely life-destroying
for a patient.
A special remark about countertransference belongs in the
hazy area of rationalization. The theoretical model one uses in the
treatment of the borderline patient or the narcissistic personality

disorder has to be adopted on the basis of a choice, for a variety of
theoretical models of all sorts is available. To be perfectly consistent
with the psychoanalytic orientation, we have to assume that the
choice of theoretical model, whether it be behavioral, supportive,

medical, whether it be based on the work of Kernberg, Kohut, Zetzel,
Klein or whomever, is multiply determined or overdetermined by

both the therapists autonomous rational ego function and his
conflictual needs. As long as a reasonable model is chosen which has
some justification in practice, this by itself does not produce any

difficulty.
The trouble is that a theoretical model can be chosen to justify
a series of acting-out techniques in the treatment, and that is why one

must be very careful in evaluating the choice of theoretical models
and in evaluating a therapists work with borderline patients. Look at

the clinical phenomena to see what the therapist is doing with his
patient, regardless of his theoretical model. Remember Freuds
(1914a) admonition: I learnt to restrain speculative tendencies and
to follow the unforgotten advice of my master, Charcot: To look at the
same things again and again until they themselves begin to speak.
For example, the technique of Kohut, in which the idealization

of the therapist is permitted over a long period of time so that the full
transference involving the search for the idealized parent imago is
permitted to develop, can easily be used by an untrained or untreated

therapist as an excuse to permit a flattering kind of worship and to
massage the narcissism of the therapist. Conversely, the technique of
Kernberg, in which a lot of confrontation goes on with the patients

rage, can be used to act out hostility and aggressiveness and to
produce chaos, either to take it out on the patient, to discharge ones
own rage on the patient or to deliberately create a therapeutic
situation in which chaos and rage reign supreme. This would be an

example of externalization on the part of a therapist who is more
comfortable with situations of chaos and rage. Thus good, sound

theoretical models which have been carefully thought out and worked
over by highly respected and very excellent authors, psychoanalysts
and thinkers can be used by the untrained and untreated to

rationalize just about anything they want to do!
Remember, much of the negative countertransference that
arises in the psychotherapy of the borderline patient comes from

therapist discouragement (Wilie 1972), which is sometimes
concealed by phony optimism. We are dealing with a long, slow

procedure. Therapists have a narcissistic need to cure, and there is
much social pressure on them to cure. Furthermore, we are
confronted with the powerful needs and the emptiness of the patient
day in and day out. We are drained, and our own needs and
emptiness, wherever they may be in our deep unconscious, are
stirred up. On the one hand we have to avoid primary process, the
tangible and the touchable, and on the other hand we must genuinely
care for the patient.
We are in a situation where we have to be an auxiliary ego in

terms of the atmosphere we provide. We have to be very patient,
consistent and reasonable. We must not exploit the patient or
retaliate against the patient. Our ethics are constantly on the table.
There is a tendency to identify with the patient out of hatred, the
hatred having been stirred up by suicide threats or attempts and
various other provocative acting-out behavior, and sometimes then
if we are overidentified with the patientwe feel depressed, we feel

as if we are empty like the patient is empty. The typical behavior at
this point is to come with the prescription pad and to fill us both up.
How many prescriptions are written for borderline patients because
they stir up emptiness in therapists who then want to fill the patients
and vicariously fill themselves?
A similar problem causes us to look for gratification from the
patient. To sit with provocation, rage and emptiness for long periods
of time and to inevitably have our own problems and emptiness

stirred up causes an almost reflex searching for gratification from the
patient. The patient is not there to gratify us, so we withdraw our

affect in revenge or conversely we make a seductive assault on the
patient to try to seduce him into loving us and so on. It is usually a
good rule of thumb to remember that in the deep unconscious
somewhere, the intense wish to rescue the patient is the other side of
the coin from the wish to kill the patient. If the therapist finds himself
flooded with intense rescue fantasies, he better beware that
underneath is a very negative countertransference!
Countertransference crises often have their origin in whatever
role the patient has put the therapist in. Sometimes he has unwittingly
accepted this role is insufficiently understood by the therapist. For
example, the most simple, common and obvious cause of
countertransference crises is the repeated threat of suicide. Every

therapist becomes uneasy when a patient threatens suicide. He
anticipates all kinds of problems, complications, disappointments and
terrible publicity, and of course it reverberates into the very depths of
his own being to see someone with whom he has a relationship

threaten to kill himself. Often threats of suicide are used by patients
to manipulate or maneuver the therapist, to disrupt the treatment

and in many ways to punish the therapist and the patients family.
Dealing with suicide is not as big a problem if the therapist is
willing to be very straightforward about it and is very carefully aware

of his limitations. If a borderline patient (or any patient) threatens
suicide in a manner that seems on clinical judgment to be serious, it is
necessary to bring family members into the therapy if at all possible

and to inform the patient and the patients family that hospitalization
is necessary. This is usually straightforward. If the patient refuses

hospitalization when the therapist feels that it is appropriate, then the
therapist has to insist that either hospitalization take place or the
patient and family have to find another therapist. Otherwise the
patient has the therapist over a barrel, raising the therapists anxiety
level and leaving him helpless and impotent to do anything about it.
A bigger problem is the borderline patient who repeatedly

threatens suicide but demonstrates little clear clinical evidence of
suicidal intent. We obviously cannot have patients going in and out of
the hospital every other weekend, because it has a totally disruptive

effect. In such situations the patient and the patients family have to
be told straight out that there is a suicide risk with any patient who
threatens suicide, even if it seems to be essentially a maneuver of one

kind or another. It is often necessary to point out to the patient and
the family that they have to make a choice whether or not to take the
risk of having such a patient in out-patient therapy and that they have
to be aware that there is such a risk. They also have to be told that the
therapy cannot be optimally continued if it is going to be continually
disrupted by such threats, and the therapist has to make it plain that
his anxiety level is also at stake. An anxious therapist cannot hear his
(or her) patient.
One can see from this simple discussion of suicide how

borderline patients can torment his therapist if the therapist is not
clearly aware both of the level at which the patient is functioning and
of the psychodynamic meaning of the patients suicidal threats and

gestures.
The exact same thing is true about the use of drugs. Drugs have
a very appropriate place in the office psychotherapy of any patient in
terms of relieving intense suffering for brief periods when it occurs.
The therapist tends to get into trouble in two areas. These are: first,
when he uses drugs as a substitute for understanding what is going
on with the patient; and second, when he uses drugs out of his need to
cure the patient.
The ambition to forcibly cure a patient by some magical
gesture is a particularly lethal form of countertransference, especially

in therapy of borderline and schizophrenic patients (Chessick 1969).
Any therapist using drugs with patients should have a clear notion of
why he is doing it at the time and should continuously on a weekly
basis review to himself whether the drugs ought to be continued and

why. In psychotherapy of borderline patients who are not chronic or
institutional patients, prescriptions for drugs should be written for
small quantities and rewritten on a weekly or fortnightly basis. They

should be marked Not Refillable, which will ensure that the matter is
constantly brought up and reviewed by the pharmacist, if by nobody
else.
A similar situation occurs in another common problem, that of

missed sessions (Chessick 1974b). Patients use missed sessions just
suicidal threats or the abuse of drugs to torment the therapist. It is
important for the therapist to recognize that payment for sessions is
for the needs of the therapist and that therefore patients must pay for
missed sessions unless in the therapists judgment there is a good and
fair reason for the session being missed. This always leaves an option
for discussion. The therapist should never be rigid. One should not in
psychotherapy try to force a patient to pay for sessions missed for a
good reason, such as acute illness, or even in typical situations where
a patient is married and the spouse has a vacation and it is the only
vacation time available for them together. A benign attitude toward
the patient is more important than anything else. In psychotherapy it
is better to err in the direction of being taken advantage of than to err
in the direction of being too strict, but missed sessions and also lack
of payment of the bill should not be allowed to be used as a way of

manipulating or tormenting or punishing the therapist. As one gains
experience, the judgment of what is legitimate and what is a
manipulation becomes easier; in the meantime, consultation is often

helpful.
In general, patients use overt rage, the creation of uproar,
acting out in the transference through erotization and seduction
involving others or other forms of acting out as sensory-motor

patterns (Piaget) of communication. It is relatively easy for the well-
trained and well-treated therapist to deal calmly with this uproar, as

long as he doesnt have too many narcissistic ambitions of his own.
The principle is, of course, to point out to the patient that the uproar,
rage attacks, etc. disrupt the treatment, make it impossible for the
patient to listen to what the therapist is saying and in some cases
given the mores of the society in which the patient livesget the
patient in trouble with the law and even get him jailed. A great deal of
the uproar and acting out can be stopped by pointing out to the
patient that you dont make calls to the jail or, if the patient is riding a
motorcycle and not paying attention to the street signs, even to the
hospitals fracture ward. The goal is calling the. patients attention to

the vital recognition that he is disrupting his own treatment.
Two of the most difficult problems are the erotization and
seduction of the therapist on the one hand, (Chessick 1966) and
chronic rage (Kohut 1972), aimed at the narcissism of the therapist,

on the other. Both of these are more difficult because they are often,
in an unconscious fashion, calculated and parceled out in small

amounts so they are not obvious and overt! Thus, for example, if a
patient comes in, sits on the chair and pulls up her dress, that is
obvious and overt, and any therapist who deserves the name of
therapist would tell her to pull her skirt down and then discuss the
meaning. But what about the patient who sits in just such a way that
at just a certain angle a young male therapist can see up her dress and
who in a thousand little ways indicates to the therapist how
wonderful he is and how romantic she feels about him?
The well-trained therapist can relatively easily accept,
confront, and interpret overt narcissistic rage attacks, and each

therapist has pretty much his own style of doing so. If the patient is
amenable to insight, the psychotherapy can then proceed. The

expression of narcissistic rage actually gives relief to the patient. In
addition, if the therapist can stand his ground and not be
steamrollered by this narcissistic rage, the patient becomes able to

gradually incorporate the therapists way of dealing with the patients
rage.
Much more difficult to manage is the chronic rage that patients
have because they are unable to get from the therapist all the
gratification of their wishes for the ideal parentthe constant little
pricklings that the patient produces hour after hour when he spots
the minor narcissistic weaknesses of the therapist. The clever patient
who has especially calculated intuitively the therapists narcissistic
weakness, who complains in little ways of the expense of the
treatment, of the idea of the treatment, of how little he is getting out
of the treatment, gradually wounds the vanity of the therapist. I
believe it is this kind of atmosphere that gradually shifts the
therapists benign attitude toward the patient to an increasingly

aggressive and sadistic one.
This is often the point at which the therapist begins either to
badger the patient, to force interpretations, to exhort, to advise, to

sermonize or to come out with a prescription pad to give the patient
drugs. It can have a more lethal outcome if the therapist actually
begins to manipulate things to get rid of the patient. For example, a
patient with considerable narcissistic rage and years of experience in

psychotherapy was gradually able to make a therapist feel that he was
not doing a good job and was not sufficiently responsive to the
patients needs. One day the therapist had to miss a session and
rescheduled the patient for a different day at a different time. When
the patient arrived, he found that the therapist had also scheduled
another patient for that time. This was discussed at some length but
not in depth, and the issue was passed over until a similar incident
took place again a few months later! In giving his vacation dates to the
patient, the therapist gave the wrong date of return, so that the
patient came down to the office a day early and found an empty,
closed office. At this point a consultation became necessary, because
the therapy went into a complete standstill.
Another typical maneuver at this point is for the therapist to
decide that the patient is not amenable to psychotherapy and thus
should have a less frequent supportive treatment, say, for example,

once every two weeks or once a month, or perhaps a group therapy;
this is duly recommended to the patient. A related

countertransference maneuver arises when the patient begins to miss
sessions and the therapist decides that this means that the patient is
getting too intensive a treatment, and he then allows the patient to

determine the frequency of the psychotherapy. This is acceptable
technique with very anxious patients who cannot stand too much
frequency, and of course the therapist has to titrate the frequency of

the sessions according to the anxiety level of, for example, adolescents
or patients who have a homosexual-panic problem. To do this with a
patient who is obviously manipulating or who has found that the

therapist gets irritated or enraged when the session is missed is
losing sight of what psychotherapy is all about.
The most useful indicator that something has gone wrong in

the psychotherapy is when decisions by the therapist such as
changing the frequency of hours, what to do about threats of suicide,
the giving of drugs, or the handling of missed sessions produce

further uproar and chaos which last for a substantial time and lead to
further such destructive decisions. It is evident that a transformation
has taken place from a healthy and therapeutic relationship to one in

which both parties are trying to protect themselves from the
consequences of their own anger and the frustration of their own
narcissistic needs. This is the time when consultation is extremely

useful, and it can often put the therapy back on the right track.
This is also often the time when consultation does not take
place, because the therapist is too angry and may even be ashamed of
his anger. Unless he is aware of what is happening, he tends to simply

act out to get rid of the patient.
The erotized transference and countertransference and the

raging transference and countertransference are furthermore related
to each other in a very important and complicated way. Blum
(1973)explains, for example, how a therapy can be silently stalemated

by an unconscious conspiracy of mutual admiration and endearment.
He sees this as a subtle repetition of the parents use of the child for
their own narcissistic needs. Thus, countertransference can divert the

tensions of transference into shared erotic fantasies or frightened
flight. Confronted with the calculated narcissistic rage of the patient,
this tension can be diverted by the therapist into his falling in love
with the patient, becoming preoccupied with erotic fantasies about

the patient, or taking flight from the patient. At the same time
countertransference behavior can anchor the patients fantasies into

transference reactions in a reality of actual seductive responses by
the therapist. This similarly leads to a deadlock.
Every therapist must arrange in advance (1) that he has a good
consultant available and (2) that if he finds himself falling in love with
a patient he is determined to go to that consultant. The reason for this
vital advance program is that when the therapist actually begins to
experience the feeling of falling in love with the patient, the greatest
temptation by far is to do nothing. Often it is a very pleasant
sensation; it is often ego syntonic, and having been rationalized, it
even seems to make sense sometimes! The patient is invested with a
certain beatific and erotic radiance which is characteristic of all falling
in love. Because that radiance appears or is perceived by the
therapist, it seems to justify his falling in love.
The proof that this is countertransference is that if
consultation takes place and correct analysis of the
countertransference occurs, the radiance will (often suddenly)

disappear and the therapist will find himself wondering, Why on
earth this particular patient? It will be almost as if a sheen or a halo
or a background light has suddenly been put out in back of the

patient.
When we find highly erotized transference or

countertransferences or highly raging transferences or
countertransferences, we must examine the material meticulously to
see what is a defense against what. Careful examination and

understanding of the patient and the psychodynamics will again and
again clarify the situation and lead directly to the solution. This
cannot help but be therapeutic, because direct confrontation and
honest discussion is the exact opposite of what the patient has

experienced as a child. In those situations everything was hypocrisy,
everything was covert; everything was usually distorted to make the
parents look good and the child look as if he were at fault. Even reality
testing and reality perception were deliberately distorted! The
parents manipulated so that the child was made to feel that the
parents were good and he was bad, and the child was used for the
narcissistic needs of the parents.
Most of the maneuvers patients use that lead to these
deadlocks are attempts to find out how the therapist is going to relate
to the patient. Will he deal with these things in a straightforward,

honest way, thus providing a new model, or will he be a repetition of
the disappointing and destructive model that the patient experienced

from the parent? Giovacchini (1975b) reminds us, The patients
defenses attempt to create a situation in treatment designed to force
the analyst to abandon the analytic role. .. . The patient has suffered
all of his life. Is it surprising, then, that he wants us to suffer for him?
Gradually we learn to absorb the patients suffering without feeling
too uncomfortable, a discomfort which is mitigated by our witnessing

the release of the patients developmental potential and the gradual
emergence of his autonomy.
Countertransference does not have to be a problem in
psychotherapy. It is the therapists attitude toward

countertransference and what he does with it that determines
whether it will be a tremendous hindrance, even destroying the

psychotherapy, or it can even be helpful in obtaining more insight
about the patient and oneself. Furthermore, one may argue that an
affirmative or therapeutic form of countertransference is a necessity

if psychotherapy is to succeed. Spitz (Gitelson 1962) introduced the
phrase diatrophic function of the analysthis healing intention to
maintain and support the patient. Thus, in response to the patients

need for help the analyst offers an empathic imbrication with his
patients emotions that provides a sustaining grid of understanding
(or resonance). The result is a certain rapport or alliance, which

leads the patient to a new beginning.
The diatrophic position of the therapist arises as a response to
the patients anaclitic regression and represents a regression by the
therapist in the service of the ego. The difference from active therapy
with direct intent to cure is that in the latter, direct libidinal
gratification is deliberately provided; the active therapist suggests

himself as a substitutive good object. In contrast to this, the
psychoanalytically informed attitude operates as an auxiliary to the
patients ego with its own intrinsic potentialities for reality testing,
synthesis, and adaptation (Gitelson 1962).
Chapter 22
Helpful Clinical Suggestions
The emptiness of the therapeutic relation is a very important

problem in the treatment of borderline patients. For example, those
with the narcissistic personality disorder who are using the therapist
as a self-object dont really relate to the therapist as a human being
with a self of his own. This is bound to stir up countertransference.
Treatment of most borderline patients, whether a couch is

being used or whether it is a face-to-face treatment, is almost always
a psychoanalytically informed psychotherapy. Parameters have to be
introduced; for example, the therapist has to be somewhat more

active and cannot tolerate many long silences. A borderline patients
silently lying on the couch for a long period of time indicates

something is seriously wrong, and it is very dangerous to sit passively
while this happens.
One must actively inquire, when one is dealing with a

borderline patient, about what is going on outside of therapy hours.
We have to ask! We dont wait patiently to find out, because by the
time we do find out, irreversible disasters may have taken place. After
a while the patient gets the idea that the therapist is interested in
whats going on in his real life. This by itself constitutes a salutary
form of limit setting. Conversely, if no questions are ever asked about
the reality of the patients life and if one concentrates solely on
dreams, fantasies and free associations, the road is paved for a
disaster in the patients external life.
It is debatable in the psychotherapy of the borderline patient

whether a true classical transference neurosis forms. There is no
doubt that a workable transference relationship often does form, but
one gets into sticky metapsychological terminology when one tries to

distinguish between a workable transference and a true transference
neurosis. The therapist must have a great deal of flexibility if he
works with borderline patients. He must learn to suit the treatment to
the patient, not the patient to the treatment. If a borderline patient is

advised to lie on the couch four times a week, sometimes the therapy
will go forward and will work, but sometimes the patient cant stand
it. The deprivation of not seeing the therapist or even the lack of
structure is absolutely intolerable. To insist that the patient use one
rigid procedure, regardless of what it is, is a great mistake in the
psychotherapy of the borderline patient (or any patient).
We want as smooth and workable a relationship as possible.

We meet the patients needs tacitly by being able to listen and by
demonstrating that we have empathy with the patients fears and
anxieties. This is the sharpest contrast we can personally present to
the parents of the patient, who could not listen to the patient and who
could not empathically grasp the patients needs.
The way we structure the psychotherapy is especially

important to the borderline patient. Paradoxically he always fights
the structure, because all structures and all limitations remind him of
irrational authority and his own sense of helplessness. Therefore the
structure has to be flexible and reasonable and it has to make sense to
the patient. He doesnt take it on faith.
If one works with borderline patients, it is important either to
have medical training or to be very closely in touch with a consultant
who has it. This is because borderline patients are in serious ever-
present danger of developing psychosomatic disorders (Chessick
1972b, 1977b, 1977c). Either these appear as a result of the
regressive processes that are taking place in the patient anyway or
they are stirred up by therapeutic regression or the patient engages in

medically dangerous acting out, which can be another type of
behavior that requires medical knowledge. If the patient doesnt take
his digitalis or his insulin and gets wobbly and dizzy, the therapist has
to know that he may not have taken his insulin or (if he has cardiac
symptoms) his digitalis.
As a general rule of thumb one should never, never, never
disregard physical complaints made by a borderline patient. They

should always be taken seriously. They should not be brushed aside
as psychosomatic. They should be medically checked out, because
the therapist often does not know the source of these complaints;
there could be organic changes regardless of whether the etiology is

psychic or not. The method of checking out these complaints is to
have at hand a thorough and reliable and understanding internist to
whom the patient is referred, as well as a coterie of specialists
available for referral, such as gynecologists and neurologists that you
are used to working with and to whom you can send the patient.
It is invariably an error in the psychotherapy of borderline
patients or patients with narcissistic personality disorders to do
physical examinations on them yourself. This produces an entirely

different kind of primary-process interchange and gratification that
will encourage the development of more psychosomatic symptoms in
order to produce more physical examinations.
Borderline patients often present life-and-death issues
dramatically and quickly, including even such issues as whether to kill
or be killed. Sometimes their regression is disruptive and it is up to

the skill of the therapist to deal with it, to know how much of it to
encourage in the psychotherapy or how to vigorously oppose it if he

possibly can. Obviously, if a patient is inevitably going to regress, he is
going to regress, no matter how hard the therapist works against it.
Therapists who are puritanical, extremely overconventional,
and prone to sermonize about ethics and sexuality do poorly with

borderline patients. Rigid morality, rule giving, or condemnation of
patient implies a grandiosity and omnipotence on the part of the
therapist that borderline patients cannot tolerate. Furthermore,
because of their wobbly ego function, borderline patients get into

many kinds of bizarre scrapes and indulge in all kinds of sexual
behavior which an overconventional or puritanical therapist may find
repulsive, disgusting and unacceptable.
It is important to keep in mind that the therapists
countertransference should not be confessed to a patient. There are
some papers and books in the literature which advise one to share
countertransference with the patient. This is a beginners mistake.
The patient does not come to therapy and pay to listen to the
therapists problems! When one has countertransference problems,
as everyone does, one resolves them, after promptly identifying them,

either by oneself, through supervision, or through more therapy for
oneself. One uses the countertransference problem as much as
possible to understand more about the patient. If one gains any

understanding of the patient, one reflects that back to him without
having to discuss with him ones own problemsthis is exploitation!
It is important to keep in mind that the use of medication with
borderline patients has a similar kind of danger to doing physical

examinations with them. There are times when it is inhumane in this
day and age not to give a patient tranquillizers or even help with
sleep, and it can be a manifestation of countertransference to

withhold medication in emergency situations as well as to give it out
all over the place. If one is reasonable and humane and very careful,
there is a place for both the neuroleptic and the anxiolitic as well as
the soporific drugs in the treatment of the borderline patient. It is
incumbent on the therapist to be knowledgeable about these classes

of drugs and to know how to use them when they are indicated. A
good rule of thumb is that for the majority of borderline patients it is
not necessary to write prescriptions except on rare occasions. The

number of prescriptions the therapist writes per week for his total
borderline practice should be small. If it is large, he is not doing a
good job of psychoanalytically informed psychotherapy with
borderline patients. Very few prescriptions for borderline patients
are written each week even in full-time practice, but one must be
willing to write a prescription if it is indicated.
The same applies to hospitalization. These patients do not do
well with hospitalization, especially of a long-term nature. They tend

to interact in the hospital with other patients with other diagnoses,
and they act out, make trouble and get themselves hated by the
personnel. Any hospitalization should be used the same way
medications are usedfor emergencies, for brief periods of time

when it seems to the therapist either that the patient must be
protected from destroying himself or from creating social or
economic ruin of some kind, or that he is a danger to others.
On the one hand, the family should not be allowed to pressure

the therapist into hospitalizing the patient, but on the other hand, it is
necessary to listen carefully and without prejudice or hostility to
what the family has to say, because they will give the therapist
information about the patients reality which the patient may not
provide, and it may greatly help the therapist in judging what
problem he is dealing with. A certain percentage of borderline
patients must have medication and a certain percent of these patients,

not necessarily the same patients, must be hospitalized from time to
time. In fact, failure in psychotherapy with certain patients is
prevented by appropriate medication and at times by appropriate
hospitalization.
The basic clinical question has to do with the therapists
feeling about whether these patients are immutably damaged and
defective or they can be approached and cured by psychoanalytically
informed psychotherapy. Here one must make a choice with each
patient. It is clearly a tragedy either way if we make the wrong choice.

A patient who is immutably damaged and defective who is subjected
to several years of intensive long-term psychotherapy has
experienced something very unfortunate in terms of loss of time,

money, energy and so on, and vice versa. A patient who could respond
to long-term psychoanalytically informed psychotherapy who is given
just a periodic brief visit and a medication is also being terribly hurt
and shortchanged.
In psychoanalytically informed psychotherapy of these
patients, one tends away from analyzing dreams and phantasies too
exclusively or thoroughly. Thus, although we listen to the patients

dream material, we tend to concentrate more on reactions to
everyday living, job, and family rather than going deeply into
interpretations of dream material. We stay with where the patient is
more than in a classical psychoanalysis, and we proceed more slowly

in most cases.
The word change is often a turning point in the psychotherapy
of the borderline patient or the patient with a narcissistic personality.
This word often brings forth an amazing and violent reaction of rage.
The patient does not want to change and cannot change ever. The
way the therapist reacts to the challenge that he gives the patient to
change is often one of those unwritteneven nonverbal at times
turning points that decide the success or failure of the psychotherapy
of a borderline patient. We all live in a constantly changing
environment, and change is the normal adaptation of all beings. When
the patient doesnt want to change, there is going to be a
confrontation between the therapist and the patient. If the therapist is
hesitant to bring up change after that, there is going to be trouble in
the psychotherapy. On the contrary, the concept of change has to be
brought up again and again and has to be employed as a focal point

for the study of the patients life.
Why should talk of change produce such a response in these
patients? First of all there is the narcissism; why should some one
who secretly imagines himself to be perfect change? Second, there is
the great anxiety that is involved in the risk of change with a weak ego
structure. Finally, to change means to accept and recognize someone
beside oneself as a functioning personality. That is the beginning of a

healthy relation to another human being, in this case the therapist.
It takes infinite patience and a great deal of empathy on the

part of the therapist to work on the reconstruction of a very defective
ego for a long time, even after the first breakthrough on such matters
as narcissistic facades has been achieved. The reward of such patience

is to help some sick but often very valuable people, at least to
sometimes avoid a functional collapse or psychotic breakdown in
their personalities. Also our own insights grow about ourselves, and
our depth of understanding and technique in dealing with emotional
illness in general is greatly enlarged.
I believe that in the psychotherapy of the borderline patient,
there is always a corrective experience. There is a counteracting of
the deception and frustration the patient experienced in early infancy
or childhood, especially in the area of empathy. After the patients
confidence has been gained, there is a concentration by the therapist,

on the reality of the patients life, and the therapist has to be very
careful not to be fooled by lack of affect or obscure dreams, which
may mask terrific fears and terrific anxieties. After the therapy is off
the ground, concentration on change is very important.
One searches for the repressed grandiose self as it shows itself
first in the vertical split (Kohut 1971) and later behind the repression
barrier itself. The core of narcissistic fantasy around which the

patient bases his life always represents a weakening of the ego by
delusionary-type thinking, and it is very important to get at these
narcissistic configurations. This is true regardless of what theory of

narcissism one wishes to hold. Dealing with narcissistic
configurations is one of the central issues in the psychotherapy of
borderline patients, both in terms of helping the patient and in terms
of what is stirred up in the countertransference of the

psychotherapist.
Certainly this discussion should demonstrate that many
therapists ought not to work with borderline patients and patients
with narcissistic character disorders. The kind of work involved has

been described, and if one does not care to do it, one should not do it!
It is an awesome responsibility to take on the long term
psychotherapy of any patient. One should know what one wishes to
handle and get involved in, but one must also know what one should
avoid. This is the mark of any skilled and experienced
psychotherapist, and it is a basic sign of his integrity, because the
results of the treatment wholly determine the patients prognosis and
hope for a decent life. As Karl Jaspers wrote, The doctor is the
patients fate.
Chapter 23
Improvement and Repair
Clearly, the initial and basic repair that has to go on in patients

of this nature is the correction of a preverbal disaster. How does this
take place? Modell (1968) writes, A successful psychoanalytic

treatment can provide in part the experience of good enough
parental care, and an identification with the analyst can become a
permanent part of the patients ego, thus permitting further ego
maturation. In some cases the faulty, negative, or defective sense of
identity of borderline and psychotic patients can be repaired. If such a

process is successful, it leads to the development of a more definitive
self-image and the capacity to form mature love relations. In others
this does not occur. It is as if the failure of the environment at a

critical phase has proved to be decisive. Modell points out that we
have not been able to account fully for why in some cases we fail and
in some cases we succeed.
Modell has the impression that the degree of sadism and
consequently the need for talion punishment may prove to be a
determinant factor. This view is similar to Kernbergs. Patients

whose sadism is overwhelming and who do not possess some
capacity for love or for tender regard for others seem to remain
unable to take in something good from the environment. Modell

explains, They are unable to form new identifications and in a larger
sense are unable to profit from experience. To learn from others and
the capacity to love others are at bottom similar; both are based on
the capacity to identify. Without this capacity there is no possibility of
psychic growth.
This initial basic repair proceeds extremely slowly and is
characterized by often taking place in spite of what is verbally going
on between the patient and the therapist. Signs that it is taking place
can be watched for in the psychotherapy. For example, a most
characteristic sign is increase of ego span. A patient who would
explode into a variety of symptoms upon frustration shows a longer

and longer period of frustration tolerance. Sometimes the patient or
those around him will notice this and report it. It is frequently spoken
of as a softening or mellowing. Thus, if the frustration tolerance has
previously been a day, now the patient can wait a week for an
important letter or a misplaced salary check or the like without
developing the characteristic explosive symptoms.
When questioned closely as to why this sort of improvement

has taken place, the patients disappoint us. At best they usually can
give only vague answers that seem to relate to being wrapped up in

the therapist, although sometimes they even vigorously deny that
therapy has had anything to do with their improvement at all. A
therapist-patient symbiosis gradually is tacitly established in which

the patient develops an almost animal faith (in the sense that this
term is used by Santayana) in the consistency, honesty, determination

and, above all, the reliability of the therapist.
Now this introduces a somewhat different and rather hazy
aspect of the psychotherapistthe psychotherapist as a human being.
I have already hinted at this in my discussion of what happens when

the therapist confronts a patient with the need to change. Such crucial
moments can be considered points in time when the therapists

existence is shared with the heretofore isolated patient. The shared
moments represent a significant intrusion of the therapist into the
patients private world. This is very tricky. From time to time in a

well-conducted psychotherapy there is a moment of true
communication or contact between the two parties, even though such
contact may not necessarily be a loving contact. For example, it can be

during a debatea sharing not based on projective identification, but
on contact between two relatively autonomous egos. Of course, the
therapist then has to be unconditionally ready to recognize the other

person and to share his world with him.
This begins to get vague, and it is the point at which I like to
introduce residents to the concepts of existentialism. My experience

with this has been uniformly negative. The resistance to these
concepts among people with medical training in the United States is
very powerful. It is very difficult to explain what exactly is meant by

existential concepts of being there or of being with the patient in
the patients existence. Meaning is hard to point to, and unless one is
fairly steeped in the existential literature, even the words are rather
differentI-Thou relationships and so on.
Buber, for example, talks about swinging into the life of the
other (Havens 1974). Other existentialists talk about caring, staying,
presence and so on. A lot of this is close to empathy and the capacity
to really be there with the patient, and it does have some similarity to
a healthy mothers interaction with her infant. In my experience, the
expositions of existential technique in the literature have been very
unpopular with residents in psychiatry and have been rather unclear.
It is difficult to see exactly where the difference is between an
existential psychoanalyst and an ordinary psychoanalyst.
What is definitely recommended and what is important in the
psychotherapy of the borderline patient is the mental attitude that
Havens (1974) calls a keeping lookinga pushing away of every

temptation the patient offers or the therapist finds to make definite
conclusionswhich is comparable to the state of empathic listening
or free-floating attention. I would suggest that anyone who wishes to
work in therapy with borderline patients carefully familiarize himself

with existential philosophy and concepts to see if the suggestions
given by the existentialists about direct grasp of the Being or the

Existence of the patient afford a further dimension of understanding
the patient and of offering him something. Some therapists seem to
intuitively grasp the concepts of existentialism and use them; to

others they simply make no sense at all.
One must be very careful not to use the existential approach as
a rationalization for engaging in all kinds of primary-process acting

out with the patients. This is a common error and it is unreasonable.
Unfortunately, existential terminology is vague because the subject of

Existence that existentialism tries to study is also vague. Therefore
existential philosophy and terminology can be used to justify just
about anything from Christian faith on the one hand to seducing and
having sexual relations with the patient on the other hand. It does not
follow from this, however, that the existentialist has nothing to tell us.
Especially in the psychotherapy of the borderline patient, who is so
preoccupied with problems of life and death, it is necessary for the
therapist to have some knowledge of this field. I will return to
existentialism in the final part of this book.
It almost goes without saying that the demands made on the
therapist by borderline patients are tremendous. They force a great
deal of thought and introspection and brooding and reflection on the

data, and they require time for the therapist to reflect on this data. If a
therapist is trying to do intensive psychotherapy of borderline
patients, he cannot carry a tremendous practice. Its too much. They
require the therapist to avoid acting out over years of intensive

therapy during which the utmost tests are put to him to see if he can
withstand what the patient has to offer. Such incidents as the

therapists canceling or getting ill or coming late to appointments,
mistakes in the time of appointments, broken promises, vacations,
even scientific meetings, become major items for discussion in the

therapy.
There is a continuous scrutiny by the patient to see whether
these everyday matters cannot be attributed to a basic dislike

assumed to be in the therapist for the patient. Now if there is a basic
dislike, obviously the therapy will fail. It is impossible to hide this

from any patient over the years of treatment.
Hopefully, after some months or even some years of this kind
of relationship, the patient gradually swings around from oscillating
psychiatric and psychosomatic symptoms and various kinds of

bizarre acting out to behavior resembling more and more what we
see in ordinary psychotherapy situations. The therapist at that point

begins to shift roles bit by bit toward a more neutral stance, with the
aim of eliciting the basic early narcissistic fantasies that the patient
has lived around. However, this can be done only after there has been
sufficient locking in of the symbiosis between therapist and patient,
so that the patient can withstand the frustrations and anxieties
involved in the uncovering of such items as his pet and secret

narcissistic consolation fantasies and so on.
Thus, in the successful psychotherapy of the borderline patient
we utilize the transitional-object type of transference that tends to
form. If we become a transitional object to the patient, then that forms

a kind of glue that holds the patient in the treatment in spite of the
profound anxieties and the profound rages that take place during it.
Meanwhile we can uncover important material, help the patients ego
to deal with it and consequently build in defenses and mechanisms
that will promote better adaptation and better integration of such

material. The transitional-object type of transference holds the
patient in the treatment during this extremely painful process. The
therapists dedication and understanding of the importance and the

seriousness of what he is doing also holds the therapist in the
treatment during this painful process.
In those cases where a classical narcissistic transference
forms, we dont find such a primitive type of transitional object
transference, but the principle is the same. The narcissistic
transferences actually become something of a transference neurosis,
depending on how you define that term; at any rate they become
important enough to the patient that if the therapist behaves himself
and conducts the therapy properly they hold him in the therapy.
Obviously the great problem with the borderline patient is to get him
to stay in therapy and locked into treatment, because it is so painful
and generates so much anxiety and so much frightening rage that the
patient has to deal with the tendency to run away from it or to act out
massively and destroy it.
It is very important to look more technically at what we mean
when we talk about basic change accomplished by the process of
psychotherapy with the borderline patient. An intrapsychic structural

change is hopefully an ongoing process throughout the course of the
psychotherapy. It has to be considered on a continuum involving the

degree of change which is occurring and also the rate of change.
DeWald (1972) points out that change may be manifest in a
specific microstructure, such as a defense mechanism, or manifested

in a macrostructure by change in broader groups of functions
simultaneously. Structural change may involve the progressive
dissolution and ultimate elimination of specific pathological

structures. For example, one observes the dissolution of specific
defensive responses to an archaic fantasy, or further, the subsequent

establishment of new structures to replace the previous pathological
ones, such as the development of the capacity for sublimation of a
drive which has been previously dealt with by mechanisms of defense
such as reaction formation and so on.
How do we measure structural change? Clearly, two general
approaches can be employed (DeWald 1972). We can use the patient

as his own control, comparing his psychic structures at the present
moment with previous levels and types of function and even with his
status prior to the beginning of the treatment. Again, we can look at

microstructures or macrostructures. We look for changes in the
intensity, frequency, and stereotyped nature of the automatic
(structured) responses that have existed, and we look to see whether

new automatic patterns of functioning have emerged.
Another approach, which can be taken essentially at the same

time, is to assess how closely the patients various psychic functions
approach a theoretical ideal or composite image of psychic structure
regarding particular issues that are at hand. Here we compare the

changes in a particular patient against a kind of hypothetical
composite image of modes of functioning in a theoretically healthy
individual. This is very tricky, because we should not introduce our
own personal standards and values in this kind of comparison, but it

is appropriate, since we do have some notions of the general way that
a healthy individual functions as compared with an unhealthy
individual.
An alert psychotherapist uses both of these approaches

simultaneously, observing the rate and extent of changes occurring in
his patient and comparing them to the patient as the patient was and,
at the same time, comparing and contrasting them to the therapists
own concept and understanding of the potential degree, range and
rate of change that the patient could undergo. DeWald (1972)
mentions certain key indicators or signs of structural change. One

such sign is the increasing richness in the recovery of infantile and
childhood fantasies and memories and, obviously, the undoing of
infantile repression. Such memories and fantasies as they are

uncovered have a quality of immediacy for the patient. They carry a
sense of conviction and personal experience and frequently occur
unexpectedly, often accompanied by the concurrent childhood

affective experience. When this happens, it is an indication of
structural change. Another indicator of core structural change is that

once basic nuclear conflicts have been resolved, their various
derivative manifestations will change spontaneously, even without
specific scrutiny in the psychotherapy and sometimes even without

conscious effort by the patient. This may even include behavior which
is recognized as having been symptomatic only after it is modified
and disappears. In my clinical experience, the patient will report this

with a sense of delight.
Another indicator to look for is the patients reaction to
previously traumatic or anxiety-provoking material. Now he is

capable of remembering, accepting and understanding the traumatic
experiences that previously evoked intense affect. Still another
indicator of structural change is the nature of the patients dream
work. We watch for the increasing freedom and directness with which
the underlying dream thoughts and wishes can be expressed as well
as for the patients increasing ability to interpret his own dreams. The
work the therapist has to do on the patients dream material
diminishes as the patient undergoes strengthening of the ego and

structural change. Such a change reflects an ego and superego
acceptance of unconscious infantile drives and drive derivatives, and
an increasing confidence in conscious integrative processes for the

control of such drives.
Still another indicator is the changed nature of the patients
relationships with people outside the psychotherapy. The patient
begins to manifest more realistic expectations and responses toward

them and is increasingly capable of tolerating and adapting to
realistic stress and frustration in such relationships. Much of Strupps
(1973) writing about psychotherapy has to do with this indicator,

which he feels is essentially obtained through conditioning and
learning in psychotherapy, but it is also possible to see this as having
been made possible by a structural change in the ego.
Similarly, another indicator is the patients growing

dissatisfaction with previously gratifying infantile relationships or
objects and the replacement of these by age-appropriate, realistically
satisfying objects. A significant enhancement or deepening of the
patients affective life and responses occurs, particularly when his

previous patterns of behavior in this regard have been inhibited and
restricted. The implication is a change in the defensive ego structures,
a modification of superego attitudes in regard to affective life, and an
increasing confidence manifested in self-esteem and the capacity to

tolerate and manage the affects.
Genuine improvement in object relations can only begin with
such patients when narcissistic fantasies have been uncovered and
given up and the patients recognize that they can get on in the world
without them, but to get to this point sometimes requires a very long
period of intensive psychotherapy. Clinical evidence of improvement

appears in such areas as showing greater ability to empathize with
others, showing greater consideration for the feelings of others,
developing closer and more mature relationships with family and

friends, and manifesting interest in and concern for community
problems.
In addition, we aim at transformations of narcissism, as

described by Kohut (1966). Depending on ones theoretical
orientation, this is an additional indicator for structural change. If we
believe that narcissism has a separate line of development from that
of object libido, as does Kohut, then we aim in our psychotherapy

with the borderline patient at what he calls transformations of
narcissism. According to this theory we are less hopeful that there

will be a tremendous outpouring of improved object relationships
because of the fundamental defect in the patient, and we are more
willing to settle for what Kohut would call healthy narcissism.
Technically speaking, we hope that the grandiose self will

become gradually integrated into the web of our ego as a healthy
enjoyment of our own activities and successes and as an adaptatively
useful sense of disappointment charged with anger and shame over

our failures and shortcomings. Similarly, the ego ideal may come to
form a continuum with the ego, as a focus for our ego syntonic values,
as a healthy sense of direction and beacon for our activities and
pursuits, and as an adaptively useful object of longing
disappointment, when we cannot reach it (Kohut 1966).
Clinically speaking, we assess the healthy transformations of

infantile narcissism through the process of psychotherapy and watch
for evidence of the egos capacity to harness the narcissistic energies
and transform the narcissistic constellations into more highly

differentiated new psychological configurations. We look for
creativity, the ability to be empathic, the capacity to contemplate
ones own impermanence, a sense of humor and, finally, the

attainment of what Kohut calls wisdom. Kohut defines wisdom as a
stable attitude of the personality toward life and the world, an
attitude which is formed through the integration of the cognitive
function with humor, acceptance of transience and a firmly cathected

system of values.
Tolstoys War and Peace, which is not mentioned to my

knowledge anywhere in Kohuts writing, is an enormous novel. If you
study and read it closely, in many ways it is the story of the
transformations of narcissism in all the important characters, five of

them; Pierre, Prince Andrew, Nicholas, Natasha and Marie, exactly
along the lines described by Kohut. The suffering and vicissitudes all
five have to go through as they mature through these transformations

of narcissism present a magnificent literary description of what Kohut
is talking about.
Thus we hope in the psychotherapy of the borderline patient
that narcissism will be transformed and reshaped into aspects of
wisdom as a result of the therapy. This hope stands in contrast to the
classical expectation of a more dramatic change from narcissism to
object love. We would certainly welcome the development of greater

and greater capacity for object love also, but we are assuming in this
theory separate developmental pathways for narcissism and the
object libido. Ideally, we would hope to see in the borderline patient

both narcissism and the capacity for object-love appropriately
transformed through the process of psychotherapy. The more
metapsychological understanding we have of the given borderline

patient, the more hopeful we become for change and improvement
and health in that patient, and the more capable we are of caring and
patiently staying with the patient over a long period of intensive
uncovering psychotherapy.
Part VI
METAPSYCHIATRY
Chapter 24
Concepts of Cure in Intensive Psychotherapy
If one looks in standard textbooks on the subject of

psychotherapy, one finds an amazing lack of discussion of the
patients basic cognitive capacity to understand what the therapist is

talking about. It seems almost incredible that so little attention has
been paid to the fact that many patients are developmentally arrested
or have regressed to earlier phases of cognitive development and
therefore they are literally either unable to find the words to describe
their feelings or to communicate their feelings to another person, or

they are simply unable to understand communications that one
would expect, if given from one adult to another, they ought to be able
to understand.
Piaget (Pulaski 1971; Evans 1973; Piaget and Inhelder 1966)
describes phases of cognitive development very carefully. The basic
developmental division is between what he calls the Sensorimotor

Period and the appearance of what he calls thinking operations. The
Sensorimotor Period lasts for about the first two years of life. At that
period of life the infant is locked into egocentrism, the lack of

awareness of anything outside the realm of his immediate experience.
Piaget describes psychic adualism as prevailing at this timethat is
to say, there is no boundary between the self and objects. Even object
permanence, the notion that something exists when you cant see it,
doesnt occur until about eight to twelve months. However, logical
structures do display themselves in the Sensorimotor Period, but they
dont display themselves through words and symbols. This, by the
way, is one of the strongest arguments that adult logic and
mathematics are not simply derived from language, but rather come
from something prelinguistic.
It is only around eighteen months of age that what Piaget calls

the Period of Representational Intelligence begins. The first early
beginning of thinking, or symbolization, or concept formation, begins
to occur around the middle of the second year of life, and with it is
brought a tremendous hunger for names. To put this another way
which is pertinent to psychotherapy, the child of eighteen months
already has a number of perceptual concepts but doesnt have the
names for them yet. This is in contrast to what was originally thought,
namely that first the child learns the name and then he gets the
concept. Reality, as we understand it then, consists of various

conceptual patterns that we build up. It is not a given; it is something
that is constantly being created every minute of our lives, and it
depends on the people we interact with.
Before the child is one-and-a-half to two years old, a form of
cognition or intelligence goes on without representation, concepts, or
symbols. From about the second to the seventh year of age occurs
what Piaget calls the Preoperational Period of Representational
Intelligence. Here begins the use of language and mental images, and
thus instead of a sensorimotor activity such as grasping, the infant
can ask for things. The mental level is still that of egocentrism and
rests on the notion that everything is made for man and for children.
It is the phase of magical omnipotence and animism.
During this phase something else very important to
psychotherapy occurs, what Piaget calls interiorized imitation. The

child is able to imitate significant adults without the significant adults
being present and actually doing something in front of him. Before
two years of age, the adult in the presence of the baby may scratch his
head and the child may imitatively do that, but after two years of age
the adult doesnt have to be right there doing it. This indicates that an
evocative or representational memory is present after eighteen

months to two years of age.
This phase is followed by what Piaget calls the Period of

Concrete Operations, from about seven to twelve years of age. These,
of course, are the grade school years, in which the child can operate
on and talk about concrete objects or their representations. Finally,

something else extremely important, the Period of Formal Operations,
begins from about eleven or twelve years and does not finish until
about fourteen or fifteen. If this cognitive period is not completed, the
child cannot go through adolescent development. This cognitive

period consists of the capacity to make abstractions, to find laws, to
think about thoughts; operations of the second order are involved
here. It enables the subject to free himself from the concrete, and it is
a very important task of preadolescence or early adolescence to reach
this point. I think one of the best ways to test whether the adolescent
has reached this point is to see how he deals with basic algebra. Can
he jump for example from the concept of number to the concept of a
letter representing a number? Then can he operate with the letters
and convert the solution back into numbers?
Whether Piaget is describing the developmental stages as they
occur in the organic maturation of the central nervous system or not

is really a different discussion. The movement from one phase to the
next is a combination of innate maturity and social stimulus in some
way, and the self-image of an individual has a lot to do with how much
success he has in getting through these cognitive phases!
Piagets own words are also interesting here on the subject of
behaviorism. His research indicates the relative unimportance of

outside stimuli in development and change in individuals and the
very great importance of what he calls internal reinforcement. Self-

regulation and internal reinforcement as they develop through these
cognitive stages become increasingly more important than external
stimulation. In general, the factors in cognitive and affective

development, according to Piaget, are a combination of organic
growth, exercise and acquired experience in the actions performed on
objects, social interaction and transmission, and equilibrium
involving self-regulation.
Now this is no simple academic matter for psychotherapists,

because we would expect in many conditions where there has been
developmental arrest to see these phases gone through as therapy
moves forward. Thus we expect to see the period of equilibration and

then a period of puzzling, of searching for new and better
adjustments, and of striving to achieve a balance between past
experience and present stress. As Piaget explains, when equilibrium is
established in one area the restless organism begins to explore in
another. Notice then that far from being black boxes or tabulae rasae
or empty organisms reacting passively to stimuli, human beings are

active explorers adjusting to the world as they find it and modifying
the world and their perception of the world to meet their needs.
Rather than passive reaction, the process is much more what Piaget
calls assimilation. Stimuli and responses interact with each other;
action schemes or thought operations are modified and enriched by
the stimuli, which at the same time are filtered through previously
formed conceptual patterns. At the very early stage of life, the
sensorimotor stage, this leads to a construction of reality in the child

that is lasting. This is the foundation of Piagets notion of genetic
epistemology.
It is interesting that dreaming for Piaget is thought of as falling
back on preoperational thought and as resembling play.
Unfortunately he uses the term operation to mean thinking in the
usual sense, and thus the term preoperational thought has to do with
conceptual operations before the age of reason, before around seven
years old or school age.
Thus it is not surprising to find a patients entire perception of
reality changing and shifting over a period of successful

psychotherapy. The most obvious example of this is the quasi-
paranoid patient or the patient who externalizes, as I have described
it in my (1972b) paper on externalization and existential anguish. The
patient may begin therapy with the feeling that everyone around him
is out to take advantage of him, to cheat him, gyp him, rip him off, and
if the patient is paranoid enough he may even believe that everyone is
out to hurt him or knife him. The patient sustains his beliefs by
selectively picking out aspects of the external environment which fit
the predetermined conception and by forcing the perceptions into a
pattern that fits a predetermined schema.
Of course, the purposes of this schema may be multiple. For

example, it may protect the patient against a surprise attack which he
has come to fear more than anything else. It may represent a
projection of the patients own hatred, which he then has to

rationalize by pointing to what he thinks he sees in the outside world,
etc. The point is that argument with the patient about what he is
perceiving gets nowhere because he only perceives what fits into the
preexisting map and this map is necessary for defensive purposes. It

cant be expected to drop away until something has happened in the
relationship between the patient and the therapist so that the patient
is more comfortable and doesnt need the map anymore. I cant think
of any better proof of how completely dependent our concepts of
reality and what we think we perceive are upon our preexisting

psychodynamic and affectual states. What a gross error it is to think
that cognition is a kind of autonomous independent function not
substantially affected by emotional factors, conflicts, etc. To think that
the individual is a blank slate that experiences can then shape in an

operant-conditioning manner is impossible, for its converse is easy to
observe clinically.
It is most important for the psychotherapist to have this all in
mind. We must always ask ourselves, At what level of cognition is the
patient and where do we hope to bring him cognitively? Otherwise we
will find ourselves speaking to a patient who does not exist, talking to
a level he does not have and thereby simply being either completely
ignored or misunderstood.
Langer (1942) points out how much more complicated our
thinking and language really are than we tend to assume. Mental life
is much more than simply discursive reason. She distinguishes
between discursive symbolism or language, the usual notion of
thought and ideas in the intellectual sense, and what she calls
presentational symbolism. Presentational symbolism includes the
nonverbal representations, the connotations, the inflections, the voice
emphasis, etc., and forms a very important vehicle of meaning, much
widening our conception of communication and having an important

influence on what the patient really gets from the therapist in the way
of communication.
If we are not aware of this we may not realize, for example,

that we are saying one thing and communicating another. This is well
known, but what is sometimes not so well known is that when we are
talking, what the patient is really listening to or looking for is
something on an entirely different level, for instance the inflection of
our voice, or the look in our eye, or what have you, and that is really
what is affecting the psychotherapynot our fancy formulations or
what we have to say! This is why for many patients the therapist may
think he is following a pure psychoanalytic model, but what the
patient is really responding to is something entirely different. It
doesnt matter that the therapist is pleased that he has made a correct
interpretation; what matters and what the patient is responding to
depends on what level the patient is at and what is he perceiving.
To put it another way, what does the therapist mean to the

patient? Goldberg (1975) points out that no therapist should fail to
ask just what role or relationship he or she has been assigned and is
performing at various points throughout treatment. More often than

not the answer will be that of a narcissistic object (a functional part of
the self) and the issues will revolve around handling of grandiose
ambitions or the yearning for powerful ideals. It is very important

for the therapist to be aware of what role he is playing for the patient
(whether he wants to play the role or not) and how the patient
perceives him. This perception in turn is based on what cognitive
level the patient is at.
Very often the relationship between the patient and therapist

gives us crucial information on the patients self-esteem and
narcissistic problems, and clues in the patients behavior are used as
communications and help the patient to see why he is disappointed in
relationships elsewhere. This is utterly lost if the psychotherapist

attempts from some fuzzy-minded notion of the psychoanalytic
model to play the role of the impassive neutral withdrawn
psychoanalyst who is unresponsive to the patients behaviora role

that no experienced psychoanalyst would present either.
This discussion also shows why it is so exceptionally
destructive to psychotherapy if the therapist practices hypocrisy. The
therapist who says, for example, that there is something therapeutic
for the ordinary patient in paying the fee, implying that paying the fee
is primarily for the patient and not for the therapists needs, is already
starting with hypocrisy. Similarly, any kind of deals that one makes
with the patient, getting anything sub rosa beside the fee, begins to set
up this hypocrisy with double messages. One cannot then know what
the patient is responding to. Therefore one cannot understand what

to correct.
The patient comes (out of the transference) into
psychotherapy seeking another chance and looking for all the love
and the gratification of infantile wishes that he never had. He is

actually putting these wishes onto an unsuitable object. The therapist
is not going to be the patients mother and gratify all the patients
needs. The way the therapist deals with the transference, especially
with the patients attempt to get all the things he didnt get as a child,
will either help the patient to integrate and mature or set up a
hypocritical situation in which the patient will be doomed to
experience another bitter disappointment.
In psychotherapy we actually have to apply our knowledge of

cognitive development into helping the patient put these vague
wishes and desires from infancy into words and concepts. Similarly
we help the patient put sensorimotor patterns into words and

concepts. The purpose of all this conceptual elaboration is to describe
what one is doing in order to gain more control over it, to refine it and
to teach it to others. In psychotherapy we try to carry the patient up
from the sensorimotor levelfrom the level of behavior that caused
him all kinds of difficulties. Such behavior is essentially
communication but not verbal communication, for it is not
conceptualized yet. It therefore is very important to urge and to help
the patients to describe their feelings, to label their feelings and to
think about their feelings, rather than to allow them to act out the
feelings or communicate the feelings in sensorimotor patterns.
By closely watching the relationship and the behavior of the

patient during the therapy, one can point these things out to the
patient and put them into concepts and words. This has the important
effect of allowing the normal developmental trends which were

arrested to take over again. Similarly, during periods of regression in
psychotherapy, the alert psychotherapist watches for the cognitive
level the patient is at and tries to gear the interpretations to that level
of vocal tone and of concreteness in speech. Otherwise, in a sense his

interpretation falls on an unreceptive brain, and no matter how
correct it may be, it is rendered inappropriate by the communication

level he uses.
The way the therapist spots these problems is by constantly
asking himself, How does the patient perceive his own behavior?
What needs is the patient trying to fulfill? How does the patient
perceive the problem, and why? It is permissible and even desirable,
if the patient is at the appropriate level, to allow the patient to cling to

the perception of the therapist as need-fulfillingproviding the
therapist does not actually make an attempt to force such a

perception by acting out and trying to fulfill the needs! In due time, as
the patient proceeds in development, the perception of the therapist
as the need-fulfilling, gratifying, all-powerful giving parent will drop
away by itself.
One of the ways this perception drops away is when the
therapist begins to be belittled and laughed at by the patient. A

perceptive therapist learns to accept this belittlement at the
appropriate time. It is very much the way that an adolescent in early
adolescence separates away from the longing for the ideal parent by
finding various little weaknesses of the parents, caricaturing them
and snickering at them and teasing the parents, etc. This is an
important form of evaluation and separation, and the therapist should

not respond to it as a narcissistic blow.
Our task in psychotherapy, very often especially at the

beginning, is to get the patient to give conceptualization to
nonsymbolized affectual or sensorimotor experiences. The way this is
done is very simple. The therapist has to do it for the patient. Very
often he has to give a verbal description to the patient of the patients
behavior in order to provide the patient with the words and concepts
to talk about, rather than allowing the patient to act out or to present
communications in a sensorimotor form. The patient in a sense, then,

is literally dragged up from a sensorimotor level by the therapists
insistence on labeling and talking about feelings and communication.
Notice that this leads to a certain model of the psychotherapist
which is different than the model of the classical psychoanalyst sitting

neutrally behind the patient. Please notice that I am discussing
intensive psychotherapy, not classical formal psychoanalysis.
The psychotherapist has a model-building function for the
patient. In treatment we find out what the previous maps and
schemas have been or what models the patient has used. Often these
are self-defeating. We try to help the patient build new ones. In terms
of their inner speech, patients often talk to our presence as if they

were talking to us when they are not even in a session. They
internalize us first. They ask what would we say. Gradually this inner
speech and these models become unconscious, and they come to

function something like the program of a computer.
The model of us as the therapist, if it is correct, cannot hurt the
patient even if we make many mistakes. It is what we are like that is

really most important. If we are basically not destructive toward the
patient and dont want to see him hurt, this attitude will be
internalized by the patient, who then will not want to see himself
hurt.
No matter how ridiculous or idealizing the transferences may
be, we must present a tone of acceptance of these transferences
recognize where the patient is at rather than trying to push or force

him to be somewhere else. For example, if the patient says, You know
its amazing how we look alike! it is not clever to point out to him,
Oh, thats not true! This is experienced by the patient as a straight-

arm which in a sense keeps him from psychologically taking from the
therapist what he needs and causes the patient to withdraw, to
become arrogant and to show a grandiose self. Thus the therapist

waits until the patient no longer needs to feel that he looks like him
and this drops by itself.
When the therapist finds himself feeling that he has to push
and correct the patient, especially to correct the transference
distortion rather than wait for the patient to grow up on his own time,
we are getting into the area of countertransference. In general, the
patient gets better in psychotherapy if you dont dc too many wrong

things. If you say some right things, that also helps. Generally
speaking, the more ambitious you are in most cases, the more
counterproductive the therapy will be.
What we are hoping is that the patient will take on the
therapists way of looking at things, especially a more benign way of
looking at himself. The patient often hates himself. After he has

experienced your benign attitude toward him, not your ambitious
pushing of him, the patient then is able to modify this attitude. Thus,
in psychotherapy the therapist participates more to bring about an
existential encounter between two persons. The individual style is not

as important as the basic humane attitude that the therapist has on the
basis of his own thorough psychotherapy in working through his own

narcissistic problems.
One might say that with the psychotic patient, the matter
reaches an extreme point. Here a real relation is absolutely necessary.
The psychotic patient actually needs a real person out there to help
him in his real life, and this cannot be avoided.
As a general rule of thumb, remember that soothing rather
than raging at a child produces a better self-image. The benign
competent parent is gradually microinternalized, leaving an
accumulation of experiences to fall back on. A therapist can often tell
when he has put down or straight-armed the patient; one must not
forget that many interpretations are not heard by the patientrather
they are experienced or felt as a put-down. In those situations one

observes a developing affectual coldness; a tendency to
primitivization of thought and speechfrom stilted speech to gross
use of neologisms and so onattitudes of aloof superiority, an

increased tendency to self-consciousness and shame propensity, and
hypochondriacal preoccupations. I am of course echoing Kohut
(1971).
In psychotherapy we see the patient proceed from a first phase

of early anaclitic object choice (Gitelson 1962) to a narcissistic type of
object choice, in which he wants to be like you or who you are or who
you were or who he thinks you are, etc. Then occurs a more or less
sexualized choice, in which what is really important to the patient is
relating, not to the therapist, but to the idea that he has someone to
relate to, and finally there occurs a real object choice in which other
persons are perceived as having an existence and needs of their own.

Thus we see that before the later phase of real object love, the patient
is literally unable to perceive a therapist as a separate person with
needs. This greatly colors the way the patient behaves with the
therapist and again reverts to a problem of cognition.
To put it another way, the fabric of the ego consists of
numerous experiences that help tell the patient what to do or what
not to do in certain situations. A library of tapes of past experiences to
call on is present in everybody. Competence in human affairs has a lot
to do with this tape library. With benign early experiences a good
library exists; with bad experiences, the person is basing his decisions
on a shredded or disrupted fabric. It is our first task in psychotherapy
to build into the patient a series of benign experiences that he can fall
back on later. This is more important than deep interpretations or
bringing the patients attention to long-repressed infantile wishes, etc.
This also explains why sometimes we find patients boring.

Falling asleep on a patient is a sign of countertransference and often
has to do with our response to highly narcissistic patients who are
using us as self-objects and show no investment in us as humans, the

reaction to this is to sense the lack of investment in us and feel as
bored as if we were listening over and over to a phonograph record.
What we have to keep in mind is that our benign presenceallowing
the patient to do thisis therapeutic, and in due time the patient can
gradually give it up. Arguing with the patient or confronting the
patient repeatedly or condemning the patient for this behavior is

useless, because the patient simply does not understand what we are
talking about. He cannot perceive things any other way.
Strupp (1973, 1975), as already discussed tries to reformulate

therapeutic factors. The important factors according to him are: a
solid, reliable, and trusting relationship with his therapist, and
learning experiences in constructive living which constitute a

meaningful emotional experience.
More controversial is moral suasion implicit in the therapists
seemingly neutral, task-oriented and clinical stance. As the bedrock
of all forms of psychotherapy, the therapist establishes himself as a
good authority figure or parent and creates conditions that
maximize the chance of being listened to. He seeks to neutralize or
undercut roadblocks the patient places in the way of his teachings; he

points out maladaptive patterns of behavior and their underlying
infantile assumptions; he sets an example by remaining calm,
unruffled, reasonable and rational; he refuses to get entangled in the

patients neurotic machinations; he conveys the message that the
patient must learn to accept personal responsibility for his own
actions instead of blaming everybody else for his predicament, etc.

He teaches the patient to be less demanding and grandiose, to scale
down his expectations and to accept a more active role in managing

his life, and he conveys the philosophy of reasonableness, rationality,
moderation, mutuality and fairness as the guideposts of a good life. He
is a good kindly parent figure in some ways, and given strong

motivation to seek change (chiefly suffering) and adequate
personality resources, the idea is to win the patient over to these new
ways of living.
To put this another and less pleasant way, for Strupp the
essence of psychotherapy is manipulation in the transference, a
sophisticated technology for persuasion and influence. By taking

advantage of the positive transference, we persuade the patient to
adopt better ways of living, based on following the golden rule in
interpersonal relations. Sadder and wiser, he has been influenced to
renounce his infantile greed.
The trouble with Strupps approach is that it mixes too many
things together. How important interpretations are and how
important lessons in constructive living are depends on the kind of
patient we are dealing with. The healthier the patient, the more
intense the workable transference that forms, the more the
transference is amenable to interpretation, and the more important,
obviously, interpretations are going to be. In many borderline
patients, schizophrenic patients and patients with character

disorders, the utter lack of benign investment in the patient by
anybody makes the investment by the therapisteven his not falling
asleepand his attempt to reeducate the patient an extremely

important aspect of the treatment.
Strupp does us a service by pointing out that to try to do a
formal psychoanalysis on profoundly disturbed patients misses the
whole point of where the patient is at the time. The patient who is
that sick, who has no solid repression barrier, who cannot form a
transference neurosis, is not a patient who is amenable to a formal

psychoanalytic treatment. Saying that interpretations with such
patients are the main therapeutic influence runs the risk of an attack
by such meticulous and careful authors as Strupp. To conclude that

interpretation generally or usually is a minimally important matter,
however, is a swing to the extreme of gross oversimplification.
A variant of all this is the existential approach to

psychotherapy. Here too there is an attempt to directly influence or
heal the patient. The method used, as well as the descriptive

terminology, however, is different than either the various behavioral
therapeutic techniques or what Strupp called moral suasion. Trying
to educate and persuade, as Havens (1974) has pointed out, we find
ourselves trying to use the self. It is very unfortunate that Havens
article is entitled The Existential Use of Self, because the term self
has come to mean something entirely different in metapsychology,
and this is not the self that Havens is talking about. A better phrase
would be the use of the therapists presence.
Havens takes off from a much more extreme author, R. D.

Laing, who is among the pioneers in emphasizing the indispensability
of the therapists human presence as the essential factor in whatever
good he may do his patient. The therapist who acts as a detached

technician only reinforces his patients problem by becoming one
more in a chain of powerful individuals who have pretended to take
an interest in the patient. What is worse is the demand that the
patient too must pretend this interest is real, while all the while they
both know that the therapists response is determined by his
definition of himself as a psychiatrist, rather than by the feelings his

patient as a person arouses in him. Laing feels that faced with this
clinical detachment, the patient can only respond to what he cleverly
calls the absence of the therapists presence or, still more

destructively, the presence of the therapists absence (Friedenberg
1974). That is very important and is a good starting point to
understanding what the existentialists are talking about.
Existential therapists approach the gathering of data in
psychology by the phenomenologic method. The term

phenomenologic approach is really a very tricky one, because it is in
psychology and philosophy allegedly related to Husserls (1965)
phenomenology. Husserls phenomenology was an attempt in
epistemology to find absolute certainty in reality. That is not what the
psychological phenomenologists are interested in. The classical
example of phenomenological psychology is from Jaspers (1963), who
wrote an enormous textbook from that point of view entitled General
Psychopathology. Jaspers was attempting to directly present to us the
mental states which our patients really experience, observing them

with respect to their kind and species, carefully delimiting them, and
differentiating them by well-defined terms. A better phrase would
have been descriptive psychologythe equation with Husserl's

phenomenology is unjustified (See Chessick 1977a).
Phenomenology from the point of view of psychoanalytic
psychology is just a beginning point of clinical work, whereas from the
point of view of phenomenological psychology (Jaspers), it is the

whole of psychological work. Hartmann (1964) points out, But we
must recall that at the outset of our investigation into the scientific
status of psychoanalysis we came to the conclusion that no scientific
psychology is capable of preserving in its concepts the lived
immediacy of its primary material, and that any psychology has to

sacrifice to its scientific goal the illusion of that deeper penetration
into its subject which belongs to immediate experience. ... In its place
we gain definitive systematic knowledge.
As soon as one moves away from recording the
phenomenology of the interaction, one sacrifices vivid immediacy, or

being there, as the existentialists like to call it, and finds oneself
involved in theoretical concepts and explanations at more or less of a
remove from the immediacy of the data.
How far is removal from the immediate data justifiable?
Grinker (1975) and his co-workers feel that this removal is
permissible only to the point of forming hypotheses about the

observations which are essentially laws, descriptions, representations
and models. Hartmann and Freud did not think so. They moved even
farther from the clinical data in terms of presenting what they

considered to be causal descriptions, and in trying to do that they
introduced aspects of the mental apparatus which they conceived to
be of a causal explanatory nature. A simple example of this is the

statement that because certain thoughts appear to be either conscious
or unconscious, it follows that there is such an apparatus in the mind
as a system conscious or a system unconscious. This is the point at
which the question arises whether such concepts represent an

overextension of the evidence.
There is no way one can answer this question, because it rests

on a metapsychiatric, or philosophy-of-science, or epistemological,
premise upon which there is simply no agreement. The attempt to
abstract from the data of empathy and introspection certain

metapsychological conceptions and the postulation of concepts about
the mental apparatus such as changes and shifts of energy, etc.,
represent an attempt to bring psychology of a psychoanalytic nature

as close as possible to Newtonian physics.
The behaviorist psychotherapist rules out all methods of
gathering data except the usual empirical procedures of
measurement. The existential psychotherapist begins from a
descriptive psychology or a phenomenologic psychology and
attempts to preserve the vivid lived immediacy of the data. He
becomes engrossed in descriptions, using the language of the

humanistic imagination, that try to preserve this lived immediacy, and
he approaches psychotherapy by trying to be there with the patient
in this immediacy.
The point of the existential technique is to keep looking and
listening in a phenomenologic sense, staying right with the material
of the patient and taking everything the patient has to say at face
value rather than searching for hidden processes. This is by itself

therapeutic and is labeled the practice of phenomenological
reduction. Appearances are accepted and one tries to stay with where
the patient is, for example with how it feels to be paranoid, manic, or
simply awake. To see things as the patient sees them and experience
them as the patient experiences them is supposed to lead to a
breakdown of the doctors objectivity and authority and to the
production of spontaneous reactions which are appropriate to the

situation and are highly therapeutic, e.g. losing ones temper with a
whining, irascible, repetitious patient.
This is to be thought of, not as just opening the door to wild
behavior, but as an invitation to free expressiveness, as long as this

expressiveness serves being with and staying with the patient.
Remaining with the patient, or reaching the patient, becomes very,
very importantswinging into the life of the other, as Buber called
it. This avoids the problem of the patient suffering from the presence
of the therapists absence or the absence of the therapists presence.
The argument of the existentialist is that a distance is opened
between the doctor and the patient by the analytic technique of free
association, a gap which is unproductively filled by abundant verbal

material and abundant analytic ideas, conceptions and theories rather
than by an emotional interchange based on staying strictly with the
phenomena and the appearances that the patient presents.

Phenomenological reduction of the emotional distance between
patient and therapist is the crucial procedure, leading to a true
meeting or encounter.
It is extremely difficult to grasp exactly what the existentialists
are talking about, but I think it is well worth having a look at and
considering using, especially in those situations where there seems to
be a stalemate or a draw in the psychotherapy. I have found it useful
in such situations to sit back and to try and feel myself into how the
patient is phenomenologically feeling right then. Having done that, I
go deeper to try to find why the patient is feeling that way. This often
leads to an explanation that has been missed or even to something in
the rest of the patients real life that has been overlooked.
I think that the existentialist approach of trying to identify or

empathize with the superficial conscious feeling state of the patient at
the time is useful, but I would greatly question whether one could call
this process or the results of it a curative influence on the patient! It is

reasonable to alternatively argue that the real proof of the therapists
presence comes when he makes an interpretation. If it is a proper,
correctly phrased and correctly timed interpretation, it indicates that
the therapist has indeed been listening and doing his work of
empathy and introspection and then correctly communicating what
he has learned back to the patient. A proper interpretation at the

proper time in the proper way often is responded to with a feeling on
the part of the patient of the very strong presence of the therapist.
Carrying this discussion to a more difficult level, we may begin

with Strupps concept that the therapeutic process rests on the
development of trust in the therapists integrity. In fact the whole
procedure of psychotherapy is contingent on the development of this
trust. According to Strupp, psychotherapy can be viewed as a

technology eliminating the barriers against openness, honesty and
trust. If psychotherapy is a lesson in the development of basic trust
that unfortunately has not developed in a patient because of faulty

mother-child symbiosis, the question is What else is there? How far
can we go?
In psychotherapy what happens next is an integration, a
greater sense of identity and cohesiveness of the self, and this is

marked by generativity, as Erikson would call itan interest in
establishing and guiding the next generation, progress in the ability to
love and work, a greater expansiveness in generosity, probably a
greater optimism, altruism and creativitythe transformations of

narcissism that Kohut (1971) describes. We contrast this to despair,
in which the untreated individual is always feeling his life time is too
short and feels disgust, misanthropy, contemptuous displeasure and
so forth.
This is obviously far more complex an intrapsychic change
than any simplistic conceptions of behavior modification would
permit. By just looking at the goals of therapy one can see the
enormous difference between behavior-modification techniques and
psychotherapeutic techniques. The therapist in many ways launches

the patient on a different course of life, and he has to inculcate some
of his own values. He fosters self-examination and self-knowledge and
honesty, and he fully participates in the patients personality and

personal development.
To what extent is the psychotherapy simply an identification
with the wisdom and insight of the therapist as a model, and to what
extent are there limitations on what the therapy can do by such
modeling? I feel that there is in human life a forward force, and that
this is built into the human organism. We are limited because all we
can do as therapists is to repair and enhance ego function in the

patient by freeing the ego from internal conflicts and providing an
atmosphere in which this forward developmental force can take over

(Chessick 1974b).
This takes psychotherapy away from the realm of education,
suggestion and manipulation and places it back primarily into the
realm of evoking the patients constructive potentials. Psychotherapy

then emerges as a practice based on the assumption that many
patients have the inner motivating force to get well and heal if given
the opportunity to do so and even to go forward beyond that and

reach phisolosophical faith and transcendencewhich is beyond
what a therapist can do for a patient. If I am correct, then any idea of
psychotherapy as primarily a modeling or educational process in
which somehow the therapist influences and makes the patient do
something is a severely restricted idea, because there is a profound
unpredictability about the outcome of psychotherapy.
In the stages of life there are phases that begin with a
hedonistic or pleasure-principle orientation and then move to an
ethical or reality type of orientation. This is normally followed by the

development of a sense of identity and a sense of self, self-
authentication and self-esteem or inner sustainment, depending upon
which authors phrases you wish to employ. As we move along in

psychotherapy, taking the patient through these developmental
phases, the creative or generative capacities of the ego take over the
treatment, the therapist drops into the background, and the patient
takes over. As this happens, the therapist can no longer determine the
course to be taken. Thus, it is not possible to predict or to mold a
genuine movement to specific individual self-authentication. It is the

built-in unfolding-forward pattern in people which is their most
important healing force and also drives them on an unpredictable
course. The therapist clears away the obstacles that are present in the
patient at the level of developing basic trust and moving from the
pleasure principle to the reality principle, but if he has done his work
he should be able to observe the taking over of the creative or
generative forces that are built into the patient and carry the patient
forward into the search for self-authentication. The therapist is not
able to predict in any mechanistic or deterministic or stimulus-
response manner what form the self-authentication will take. In a way

the therapist can only withdraw and wonder at this point and watch
the unfolding of the patients personality.
Furthermore, we are really unable to predict how far the

patient will continue to unfold. I suggest that the search for
transcendence (Chessick 1974b, Chapter 15) is an even further step in
this unfolding, but this is speculative and I cannot prove it. It is

fascinating to watch this unfolding as it takes hold in our patients. We
often do not get a chance to see how far it will go, because as the
generative and forward force takes over, the patient really needs us
less and less and usually leaves us before the complete course of the
therapy is finished. I am firmly convinced that an enormous amount
of very important therapeutic change takes place after the
termination of psychotherapy, change in which the forward force

simply carries forward the development through an internal
generation. There is remarkably little research on this subject.
The crucial difference between psychoanalytically informed
psychotherapy and other forms of psychotherapy is that the former

does not directly attempt to cure, heal or influence the patient. It
works indirectly, although it recognizes that direct influences are also
at work. The degree of importance of the direct influences, whether
they be induced by therapist-presence, or education, or what have

you, is related to the state of the patients ego. The poorer the ego
function or the more fragmented the self of the patient, the more
direct pacification, unification and optimal disillusion is necessary. As
the self coheres and the ego gets stronger, a shift takes place in the
psychotherapy, allowing the patient to take over and allowing the
patients ego to pick up the psychotherapy and move forward. This is
the crucial difference between other forms of psychotherapy and
psychoanalytically informed psychotherapy, in my opinion.
Now, how do we do this? We use a special mode of observation
of which introspection and empathy are essential constituents (Kohut

1959). The limits of introspection and empathy are the limits of
psychoanalytically informed psychotherapy. This is our data
gathering, and this observational method defines the contents and

the limits of the observed field. It is not the way we try to cure
anybody.
As the patients sense of self coheres and his identity becomes

solid and his ego function improves, we shift away from direct
influencing by pacification, unification, education or optimal

disillusion (Gedo and Goldberg 1973), and we begin to listen more
and more with free-floating attention. We use empathy. We try to
understand what is going on inside the head of the patient and use
introspective self-observation as we resonate with the patients
unconscious.
In the very sick patientthe schizophrenic or the borderline
patientthe problem is different, because the therapist is
experienced as actually being the whole object which has caused the
trouble, and the patient is somehow attempting to simultaneously
separate, cling and protect himself from intrusion. As Kohut (1959)
puts it, In the analysis of the psychoses and borderline states, archaic
interpersonal conflicts occupy a central position of strategic
importance that corresponds to the place of structural conflicts in the
psychoneuroses. In these situations the real relationship, the so-
called therapeutic alliance, is extremely important, and the benign
therapists attitude or his presence (if you insist on approaching it
that way) has a major direct influence on allowing the fragmented ego
of the patient to cohere and to develop a sense of self, which is then
followed by improved ego function in many areas even without the
benefits of interpretation.
Now if that is as far as we want to go with a patient, fine, but
we may wish to try to go farther. If we do, then we begin to shift over
into a stance that enables us to reflect back to the patient what we
have learned about him from empathy and introspection. If the

patient now has a sufficiently coherent sense of self and ego strength
to hear us, this is followed by increased strength in the ego, until
gradually the internal generative force of the patient takes over more
and more, which is what we are after.
There is a long-standing feud between those who think that

there is no difference between psychoanalysis and psychotherapy on
the one hand, and those who think there is only one genuine
psychotherapy psychoanalysison the other. There is no way to
resolve this issue because it is so overloaded with status
considerations, etc., but there are certain basic clinical and practical
differences that we should be aware of. The most obvious difference

between classical psychoanalysis and psychotherapy stems from the
development of a workable transference neurosis in psychoanalysis.
If a workable transference neurosis develops and the focus of the
therapy is the analysis of this transference neurosis, then you have by
definition a formal psychoanalysis. Only a limited number of patients
are in a sufficient state psychologically, economically and
sociologically to allow this to happen.
Ticho (1970) mentions certain indicators of dangerous

transference difficulties, for example intense early transference
fantasies where the patient cannot completely distinguish between
the treatment reality and the fantasy. Beware of a lack of containment

of the regression to the therapy hours, with the transference behavior
spilling over into the outside world, and of the early appearance of
intense oral demands and masochistic wishes. Any therapist who

attempts to do formal psychoanalysis with patients while these things
develop is taking a terrible risk, requiring much special skill and
experience; switching to a less frequent face-to-face psychotherapy is
the obvious practical solution.
A variety of techniques for reducing the intensity of the

psychotic or unworkable transference are described in DeWalds
(1964) book on psychotherapy. Most important is that the therapist
must realize what is happening before it gets out of hand. I take an
essentially psychoanalytic stance with many patients at the beginning

to see how things develop. For the minority of patients who show the
potential to develop a substantial transference neurosis and who have
the social and economic capacities to undergo a formal

psychoanalysis, that is certainly the treatment of choice. Most patients
do not have this capacity, at least in the ordinary private practice of

psychotherapy, and thus most patients are seen once or in my opinion
preferably twice a week. Even this often is more than many patients
can afford.
Another important difference is in the analysts neutrality
compared to the psychotherapists. This is a much more controversial
issue. There is no doubt that the psychotherapist is a more real figure

to the patient, a reality usually facilitated by the face-to-face
relationship. In fact, one can easily argue, as I have already done, that
the psychotherapists real interest in the patient and efforts at

reeducation are tremendously important to the patientfar more
important in many phases of therapy than interpretations. There is no
agreement as to whether this is true of psychoanalysis or not, but in

general the psychoanalyst makes an effort to avoid being a real object
to the patient. Whether that is really possible or not over frequent
sessions and long periods of time is a matter of great debate.
Finally, Ticho (1970) explains that the interventions in
psychoanalysis are almost exclusively insight-producing, such as

interpretation, confrontation and clarification, whereas in
psychotherapy, interventions are also supportive or noninsight
producing, such as suggestion, advice, reassurance, persuasion,
setting examples, proving points, giving recommendations and giving

prohibitionsall of which also imply a much greater personal
involvement on the part of the therapist. The danger of this increased
personal involvement is an increased expectation of immediate

improvement.
It follows that the danger of uncontrolled countertransference
is much greater in psychotherapy than in psychoanalysis. This is a
very difficult concept and is more of a theoretical difference or one of
degree in many instances than a real and substantial one. The
psychoanalyst often attempts deliberately to keep his interpretations
as free of direct educational influence as he can, but it is really

impossible to do this in a total way, especially when the patient is
coming four or five times a week, whereas the psychotherapist feels
much more free to assert direct educational pressures, because many

patients simply need that. The techniques of pacification, unification
and optimal disillusion (Gedo and Goldberg 1973) are essentially
educational techniques which the psychotherapy patient must have to

develop a coherent sense of self.
The picture of a novice psychotherapist trying to model
himself on the neutral psychoanalyst, sitting with a schizophrenic
patient or with a schizoid or borderline patient in silence trying to

listen to free associations and make interpretations with such
patients at a deep unconscious level, represents a complete
misunderstanding of psychotherapy. In teaching residents I have tried
to compare this to a situation in which two people are sitting in a

burning house. The obvious thing to do is to get out of the house and
then put the fire out. If instead the two people simply sit around the
table and discuss the chemical nature of the process of fire, in my

opinion somebody is going to get hurt badly. It may seem strange that
I belabor this subject, but it is a fact that in many training programs
today the model of the neutral psychoanalyst is still used as the ideal
model in psychotherapy.
Chapter 25
Philosophy of Science
What are the philosophical foundations of psychotherapy? In a

field in which there is so much disagreement, it is absolutely
necessary to review the fundamental premises behind various forms

of therapeutic endeavors, for whether we like it or not we are forced
in the practice of psychotherapy to make certain philosophical
assumptions and conceptions and even forced to make philosophical
choices.
The field examining the highest or first principles that underlie
or form the ground of all scientific investigation and all thinking is

classical metaphysics, usually containing epistemologyhow we get
our knowledgeand ontologythe search for Being, or the ground of

all knowledge.
Assumptions about epistemology and ontology are at the basis
of all scientific endeavor, and since these assumptions differ from

individual to individual, they greatly influence the kind of thinking
and scientific work that takes place. From the point of view of the
psychotherapist, the most important aspects of classical metaphysics

are the assumptions behind scientific work and thought, an area
which is usually known as philosophy of science. We might
characterize these assumptions as second order principles, because
philosophy of science is based in turn on certain first assumptions
from metaphysics, ontology and epistemology.
Getting more specific, from the point of view of the
psychotherapist we have also a third order of principles, which I call

metapsychiatry (Chessick 1974b, 1977). Metapsychiatry, to be
discussed in more detail later, simply represents those aspects of the
philosophy of science that are specifically important to

psychotherapy.
From philosophy of science and me a psychiatry there is then a

jump to scientific research in areas such as physiology, brain study,
psychology andif one is willing to accept themthe research
methods of psychoanalytic psychology, utilizing data gathered by

empathy and introspection. At the very bottom of the theoretical
ladder, but closest to the actual empirical material, we have the
clinical application of these various orders of principles and of basic
researchthe disciplines of clinical neurology, clinical psychology,

clinical psychiatry and psychotherapy.
The task of philosophy differs from that of science, for unlike
science, philosophy examines not our conclusions but the basic
conceptual models we employthe kind of concepts and ordering
patterns that we use. Philosophy concerns not the explanation of this

or that but questions such as What really is an explanation? or
What really is change? or What really is a cure?
For example, Is something explained when it is divided into
parts and if we can tell how the parts behave? This is but one type of
explanation. It works fairly well for a car, although it does not tell us
what makes it run, and less well for a biological cell, the parts of
which are not alive; certainly it does not explain life, and it works very
poorly to explain personalitywhat are the parts of a person? Or,
choosing another of the many types of explanations, has something

been explained when we feel that we understand it because we have
been shown how it fits into some larger context or broader
organization?
These questions, which are essentially philosophical questions,

are not designed to determine the explanation of this or that, but to
discover what an explanation is. Yet as we have seen, there are many
different kinds of explanations. In any one case, what shall we use? Or

should we try to use them all, and, if so, when and to what advantages
and pitfalls? How is our choice among these varied explanations to be
made? Should it depend on the feeling with which we work, on what

we want an explanation for, or on the style of the times?
When we ask questions of this sort we seem to be talking
about nothing in particularsuch philosophic issues at first seem to

be empty. Yet they very basically affect whatever we study, for
depending upon which mode of approach we use, different questions
and hypotheses will be formulated, different experiments set up,

different illustrations cited, different arguments held to be sound, and
different conclusions reached! Much in our conclusions about anything
comes not from the study of the things, but from the philosophical
decisionsthe prior philosophical decisionsimplicit in the way we
start!
The whole matter is even incredibly more confused because a
number of generally held classical assumptions about scientific
method and science are clearly by this time known to be wrong. For
example, there are inductive and deductive theories, but these are not
the same as inductive and deductive inferences. A deductive theory is
not solely the product of deductive inference and inductive theory is

not solely the product of inductive inference. Bacons classical notion
that the method of inductive inference leads to the generalizations of
the inductive sciences is simply not correct!
Let us turn to classifying the various possible kinds of scientific
theories and see where the theories of psychotherapy fit

appropriately into the classification and where they have been
incorrectly placed in the past. What are various kinds of theories that
scientists work with? First come the well-known deductive theories

(which do not proceed only by deduction) in mathematics and logic.
We will not go into technical details about mathematics and logic
because we are obviously not dealing with that kind of a science when
we talk about psychotherapy.
Clearly theories of psychotherapy based on clinical empirical

material belong under the rubric of inductive theories. The inductive
theories can be further subdivided (Basch 1973) into classificatory
and explanatory theories. The classificatory theories are formed by

abstractions from observations. These are the natural sciences in
which, for example, one looks at all the birds one can find and tries to
make classificatory abstractions to divide them into groups, species,
etc. This activity forms a very important aspect of the natural
sciences. We assume that classificatory inductive theories are based
on empirical data and that the empirical work was done by a trained
observer who knows what methods to use and how to examine the
various materials. Then the trained observer has to make abstractions
from those observations.
More complicated are the so-called explanatory inductive
theories. These theories generate hypotheses about the observations,
not simply abstractions from the observations. Basically, two kinds of
hypotheses are made. First of all come laws or descriptions: How do

things take place and along what general laws or principles do they
occur? The second kind of hypothesis might be said to consist of the
causal types of inductive explanatory theories: Why do things take

place? Answers to the latter questions imply that something causes
something else.
If we look at the various forms of psychotherapy we find a
good deal of confusion about which kind of theory the author is using.
For example, take the trained observer in psychoanalysis or
psychoanalytic research. His method of observation, empathy and
introspection, is his stethoscope, his microscope, his telescope. If he is
properly free of his own problems, then he is able to listen with free-
floating attention, as Freud describes it, to the free associations of the
patient. Through the method of empathy he is then able to identify
temporarily with what is going on in the patients mind, and by
introspection into his own mind he is able to come up with
information about the patient. This is considered to be an empirical
science providing empathy and introspection are acceptable as a
method of observation. If they are not, then psychotherapy must be
dealt with strictly in terms of behavior descriptions. Introspection
and empathy are essential constituents of psychoanalytic fact-finding,
and this observational method defines the contents and the limits of
the observable field.
Grinker (1975), Basch (1973, 1975) and other authors claim
that among the inductive explanatory theories the how type of theory
(laws or descriptions) is all that is really necessary. These authors

emphasize transactions, and they use the language of general systems
theory to describe the transactions or interactions that go on between
the patient and the therapist; psychotherapy is conceived in terms of
error-correcting feedback systems. For example, much of

psychotherapy for these authors has to do with self-esteem problems
and with situations in which a patient pushes away someone whose

esteem he wants over and over again. The therapist asks, What is the
patient looking for? What is the patient doing to me? What kind of
messages is he sending? Then he confronts the patient with the here-

and-now in the transference, and in so doing he breaks up a system
which has led to great trouble for the patient.
It is not necessary in this approach to postulate concepts like

psychic energy or the mental apparatus (ego, id and superego). The
implication of this metapsychiatric or philosophy-of-science premise

is that the experienced relationship between the therapist and the
patient is the crucial aspect of the treatment in psychotherapy. We do
not interest ourselves as much, from this point of view, with
interpretations of the infantile roots of the transference. The
transference is there as a given and is utilized in getting the patient to
listen to our error-correcting feedback. In this view, formal
psychoanalysis and psychotherapy are fundamentally different.
Toulmin (1960) proposes asking for any science, What are the
methods of representation? What are the models employed in doing

so? He proposes that we look for the form of given regularities and
do not ask what is the purpose of these regularities. According to
Toulmin, science tells how things happen, not why they happen. It
consists of descriptive methods of representation. The most
interesting corollary to this, Toulmin points out, is that when two
scientists do not agree on what is to be explained, there is no hope for
their agreement on a description. Again, if the epistemological

premises with which two scientists approach the data are not the
same, it is impossible for them to agree. They will come up with

entirely different methods of representation which seem to conflict
with each other, although actually they are different because they
start from different premises. They may indeed be complementary,

and no one set of premises in the philosophy of science can be said to
be the only truth.
One has to remember that at the time Freud was writing, the
prevailing scientific attitude was quite different. Newtonian dynamics
were considered to be the answer. It was felt that the scientific

method in terms of cause and effect, with laws as described by
Newton and others, had a tremendous success and was the method.
General systems theories and other alternative paradigms did not
exist, and it was of the utmost importance to Freud, who was
founding what he hoped was to be a new sciencewhich
unfortunately psychoanalysts tend to refer to as our sciencethat
methods of representation be employed which sounded as scientific
as possible and as closely related to the concepts of Newtonian
physics as possible. Such terms as energy became very very

important, and if one looks at Freuds (1895) early unpublished work.
Project for a Scientific Psychology, one sees a kind of transition
between Newtonian science and psychoanalysis in which every effort

is made to describe the data in as physicalistic-sounding terminology
as possible. These efforts did not succeed, because the data obtained
by empathy and introspection cannot be forced into the mold of

Newtonian physics.
Freud soon gave up this approach, but his basic bias, the basic
epistemologic or metapsychiatric premise that science implies

classical mechanics, remains and is only recently being challenged.
Today it is mainly the behaviorists who are trying to adhere to as
close to a model of Newtonian science as is possible. And it is possible

to do this if one throws out all data obtained through empathy and
introspection and concentrates strictly on precise observable
phenomena of behavior.
Psychologists can further be divided into (a) those who
attempt to find descriptions or laws that describe the transactions

between the therapist and the patient in terms of the here-now
relationship and the meaning the therapist has to the patient and the
patient has to the therapist, and (b) those who make a further stepa
generalization or removal from the evidence that is even farther away
from the lived immediacyand postulate metapsychological
conceptions such as the mental apparatus, energy, instincts, genetic-
dynamic formulations, etc., which they believe provide a causal
explanation of what has happened.
These differences in basic epistemological premises explain

why it is when you get a group of psychotherapists together and
present them with the same phenomena, you cant get agreement, for
there is no agreement on what the phenomena are that are supposed

to be explained or even on what an explanation is in science or in
psychotherapy! The way out of this predicament as I see it is to use
the principle of complementarity. It is not necessary that any one
epistemological approach be the only right approach. It is clear that if

we have explanatory concepts of various abstract orders we will move
away from the vivid immediacy of the data.
Bohrs great principle of complementarity was developed in

the discipline of physics in order to meet a similar apparent paradox.
What Bohr (1934) was pointing to when he introduced his principle

of complementarity was the curious realization that in the atomic-
particle domain the only way the observer (including his equipment)
can be uninvolved as if he observed nothing at all. As soon as he sets

up the observation tools on his workbench, the system he has chosen
to put under observation and his measuring instruments for doing the
job and the observer himself form one inseparable whole. Therefore,
the results depend heavily on the observer and his apparatus.
Bohr was able to show that on the level of atomic particles any
apparatus designed to measure position with ideal precision cannot
provide any information about momentum and vice versa. Thus two
mutually exclusive experiments are usually needed to obtain full
information about the mechanical state, each complementing the

other. He expressed this conclusion as a general principle of
complementarity, and in developing this principle he asserted that it
is neither possible nor necessary to make a choice between waves

and particles, as indeed both are essential for complete
comprehension of reality.
The uncertainty principle of Heisenberg has sometimes been

misconstrued to mean a particle actually has both a precise position
and momentum until it is disturbed by the experimenter and that the
act of observing the position precisely destroys the precise
momentum. In other words it is assumed that nature is involved in a

bizarre conspiracy to prevent the discovery of something that has a
real existence. It is nearer to the truth to assert that a particle in itself

has neither a position nor a momentum and that the act of
observation creates its mechanical state.
Bohrs point of view was criticized because it seemed to treat
particles and waves as equal whereas particles are a mode of

existence while waves are a mode of behavior. But the choice is a
matter of the temperament and taste of the observer! The apparent
paradox between describing psychotherapeutic interaction in terms

of existential psychiatry or in terms of interpersonal psychodynamics
presents an identical dilemma to a participant observer. Thus, some
choose to characterize the interaction as an investigation of and an
intrusion into the patients mode of existence, whereas others are
more comfortable using Freuds psychodynamics. The point is that
both are essential for a complete comprehension of reality and it is
not a matter of one descriptive language being right and one
descriptive language being wrong.
Bohr hoped that the principle of complementarity would come

to be applied to many other areas of knowledge beside atomic
physics. As explained, for example, by Holton (1973), Bohrs real
ambition for the complementarity conception went far beyond

dealing with the paradox of physics in the 1920s: From this point of
view we realize that Bohrs proposal of a complementarity principle
was nothing less than an attempt to make it the cornerstone of a new
epistemology. . . . It was the universal significance of the role of

complementarity which Bohr came to emphasize.
In place of a precisely defined conceptual model, the principle

of complementarity states that we are restricted to complementary
pairs of inherently imprecisely defined concepts, and the maximum
degree of precision of either member of such a pair is reciprocally

related to that of the opposite member. The specific experimental
conditions, then, determine how precisely each member of a
complementary pair of concepts should be defined in any given case.

But no single overall concept is ever possible which represents
precisely all significant aspects of the behavior of an individual, for

example. The principle of complementarity renounces the notion of
the neat and precisely defined conceptual models in favor of that of
complementary pairs of imprecisely defined models and represents
an absolute and final limitation of our investigation and
understanding of every domain of knowledge.
Only those who have attended international psychotherapy

congresses can be aware of the intensity of the debate in the western
world today between the proponents of the Freudian deterministic
metapsychology and the proponents of the existentialist and

phenomenological approaches to psychotherapy. To see these as
complementary descriptions that are related to whatever critical
nuclear aspects of the psychotherapists mind being employed on the

data at the time is a way out of the dilemma. The two maps of the
reality of psychotherapeutic interaction and the two descriptions of
what is going on are not fundamentally opposedthey may be used
successfully to complement each other providing the therapist is

carefully aware of when he is using each complementary map.
Furthermore, as is the case with physical experiments, the

kind of approach, attitude and personality that the observer or
psychotherapist brings into the psychotherapeutic interaction will
determine the kind of descriptions or language that he uses to

describe the process and results of psychotherapeutic interaction.
Thus, for example, Freud, who was extremely concerned to make
psychoanalysis scientific, describes all psychotherapeutic interaction

in strict, classical scientific terminology; Buber, who bordered on the
existential mystique, uses an entirely different language in describing

the same kind of confrontation that takes place in a meaningful or
therapeutic interaction.
This notion of complementarity also explains why there is
actually not a great deal of difference between the practical technique
of psychotherapy employed by the analytically oriented

psychotherapists and that employed by the existential
psychotherapists; the enormous differences appear far more in the

language employed in describing the phenomena than in the
techniques used. Perhaps the only way to characterize this is by
describing the analogous study of a neon sign. An expert physicist

from another planet could probably describe in great detail the
electronic working of the sign, but he could never explain why people
were throwing rocks at the sign unless he knew the language of the
insulting message the sign was sending out. Similarly, in
psychotherapeutic interaction we can experience the rigid
determination of the repetition compulsion and use it in

understanding our patients behavior in one way, or we can
experience the patients freedom for existential choices and leaps into
life styles if we approach the psychotherapeutic interaction from an
entirely different standpoint.
In his famous essay The Structure of Scientific Revolutions,
Kuhn (1972) has described this shifting back and forth between
paradigms as a Gestalt switch, and what I am recommending for the
psychotherapist is clearly a similar capacity to switch back and forth
in the interest of complete understanding of the patient. I have been

surprised by the resistance to this recommendation, which usually
springs from a naive or inadequate understanding of the philosophy
of science or, more ominously, from an almost religious fervor for one
or the other paradigm. As Kuhn explains, Philosophers of science
have repeatedly demonstrated that more than one theoretical
construction can always be placed upon a given collection of data.

History of science indicates that, particularly in the early
developmental stages of a new paradigm, it is not even very difficult
to invent such alternates. But that invention of alternates is just what

scientists seldom undertake except during the pre-paradigm stage of
their sciences development and at very special occasions during its
subsequent evolution. . . . The reason is clear. As in manufacture so in

scienceretooling is an extravagance to be reserved for the occasion
that demands it.
The difficulty of resistance to other paradigms is apparent. In
fact, . .. the proponents of competing paradigms practice their trades

in different worlds. .. . Practicing in different worlds, the two groups of
scientists see different things when they look from the same point in
the same direction... . That is why a law that cannot even be

demonstrated to one group of scientists may occasionally seem
intuitively obvious to another. Equally, it is why, before they can hope
to communicate fully, one group or the other must experience the
conversion that we have been calling a paradigm shift. Just because it

is a transition between incommensurables, the transition between
competing paradigms cannot be made a step at a time, forced by logic
and neutral experience. Like the Gestalt switch, it must occur all at
once (although not necessarily in an instant) or not at all.
The principle of complementarity leads us into an overall

consideration of basic unresolved problems and limitations in our
general seeking after knowledge. The particular work of Aristotle that
was inserted by some unknown commentator after his treatise on
physics deals with these problems, or first principles as he called
them, and has come to be known as metaphysics, which means
literally after physics. The basic problems dealt with by metaphysics
are of two sorts. (1) How do we know anything?the discipline of
epistemology, and (2) What is there to know?the discipline of
ontology, or study of Being or Reality. A moments reflection will

convince you that our convictions about What is there to know?
depend on our theory of How do we know anything?
For example, the continental rationalists in the seventeenth

centuryDescartes, Spinoza, Leibnizattempted to arrive at
knowledge from reasoning or from the mind alone, and they
subsequently developed an essentially theological point of view about
what there is to know where God played a vital role. On the other
hand, the British empiricists in the seventeenth and eighteenth
centuries Locke, Berkeley, Humebelieved the arrival of

knowledge comes from sense experience alone and our mind is a
tabula rasa on which all knowledge is brought in from the outside.
This viewpoint leads to skepticism about whether there is anything

that we can know at all and to a more or less deterministic and
behavioristic approach to psychology.
The existentialists of today present a somewhat modified

version of the continental rationalist philosophy; they attempt to gain
knowledge about Reality or Being by a study of our kind of Being, that

is the Being of man as thrown into the world. For example, Heidegger
claims that the inner voice of Being can be heard in solitude if we are
authentic and do not live a life of essentially falling away from Being.
Sartre argues that crucial ethical actions and choices represent the
essence or Being of man. Husserl grounds Being in the
phenomenology of the consciousness, and Jaspers maintains that
certain boundary situations reveal man to be free to make crucial
choices. These are examples of what Jaspers calls ciphers, which
offer a chance for man to get in touch with Being.
It still seems that the best starting point is Kants combination
of (a) the sensory manifold with (b) space and time added by the
synthesis of the imagination and finally (c) the activity of the

synthesis of the understanding in producing the awareness of objects
of physics. Phenomena and noumena as described by Kant mark the
boundaries of the knowable and the unknowable. At the same time
we must constantly recognize the human tendency to try to

transgress this boundary. This human tendency could be defined as
the discipline of metaphysics, which, according to Kants philosophy,

cannot ever actually succeed in this endeavor.
Thus, human reason is ineradicably metaphysical. It is haunted
by questions which, though springing from its very nature,

nevertheless transcend its powers. We might call this viewpoint a
form of metaphysical agnosticism.
I began this section by showing how certain apparent
paradoxes arising from the clinical practice of psychotherapy of

borderline patients could not be resolved without appeal to a higher
level of theoretical discourse, namely, the philosophy of science. Thus
toward a general principle, we may state that problems apparently
insoluble and paradoxical at level L0 may be resolved at level L1; the
level Lx refers to the language employed, which always includes all

the language parts of the previous level (Lx-1) and also introduces
more abstract and meta+ language concepts.
Thus any language level Lx is a metalanguage for Lx-1 and so
forth. Generalizing, we may say that problems insoluble at level Ln

should be appealed to level Ln+ I in order to make them soluble. In
fact we may have to invent a special language L n+ I in order to do so.
This commonly occurs; the most famous example is modern
metamathematics.
However, as one ascends from Ln to Ln+1, one gets farther and

farther removed from the immediate empirical and clinical data and
closer to inborn expectations, intuitive grasps and other such
methodologies to establish principles, and so proof by empirical
methods becomes increasingly impossible. Here again, the principle

of complementarity is at work. The price we pay for resolving
problems by using ascending levels of abstract discourse is to reduce

the empirical certainty of our solutions!
For example, let us return to my argument (Chessick 1971)
that science and art are separate by their very nature and cannot be
thought of as coming together as our knowledge increases. Meyer
(1974) gives further detailed evidence for this view and then
concludes: Stern and Snow are on the side of the angels in that both
want to bring the various kinds of human activities and pursuits
together. So do we all. But mistaken analogies, however
commendable the motive behind their advocacy, will not unite
disparate disciplines or join noncomparable ways of knowing. Like
crossing a horse with an ass, they will only beget mulish recalcitrance
and sterile dispute. One way of comprehending all knowledge is
through meta-disciplines such as history and philosophy. The
difficulty with these modes of integration, at least for me, is that the
more encompassing they are, the more obscure and vaporous they
become. Instead of the clarity and concreteness of scrupulous
observation and the precision of rigorous argument, we are given

elusive spirits and untestable speculations clothed in abstruse
language.
Chapter 26
Metapsychiatry and Beyond
The concepts of science and of being a scientist have all kinds

of cultural overtones. Suppose one therapist says that anothers
approach is not scientific; that phrase has come to have a

derogatory or pejorative meaning, and therefore, each investigator
tends to insist that his method is the scientific one. This has other
overtones in terms of prestige, getting money for research, etc., but it
is very destructive to the field of psychotherapy.
Psychotherapy in general is much more complicated than
classical Freudian psychoanalysis. If one stays strictly with the

writings of Freud, and perhaps some of the extensions by Hartmann
of Freud, psychoanalysis is a very specific clear-cut method, a

subdivision of psychotherapy. If one agrees with the premises and the
metapsychological conceptions and is well trained, one knows exactly
what one is supposed to be doing. Psychotherapy, on the other hand,

is far vaguer, and many more vital factors are recognized to be at
work. This permits a far greater amount of argument and
disagreement. With psychoanalysis, one either accepts the premises

or rejects them, and that determines what follows. With
psychotherapy there are so many factors involved, so many different
kinds of patients involved, that the situation is much more

complicated.
Psychoanalysis tends to get into trouble when it attempts to
use its methods on patients for which it was not intended. Let us not
forget that Freud conceived of psychoanalysis as a method of

treatment for certain specific types of neurotics, and as far as I can see
from the literature, the only theoretically legitimate extension of

formal psychoanalysis from the treatment of the transference
neuroses has been Kohuts (1971) effort to establish the
psychoanalysis of narcissistic personality disorders as he defines

them. Even in making this effort, Kohut had to make a major
modification in psychoanalytic metapsychology.
Those who would attempt to use a classical psychoanalysis on

disorders such as the borderline state or overt schizophrenia run the
risk of introducing a tremendous confusion into the meanings of the
terms in the field, or contrary-wise they will have to make a complete
revision of psychoanalytic principles. The closest to the latter, of

course, is in the work of Melanie Klein and her followers. It is only
after one has completely revised Freuds psychoanalytic theory that

there is some theoretical justification (whether it is right or not is a
different discussion) for approaching borderline patients or patients
with overt schizophrenia by classical psychoanalysis.
In evaluating presentations of psychotherapy from various

schools, it is now absolutely mandatory for the reader to keep in mind
the basic epistemological premises of each school. If they differ from
his, he must try to evaluate the presentation in terms of the implicit

premises and not expect that the presentation will follow the lines of
his own epistemological premises. Irreducible philosophical
differences rather than right or wrong or scientific or unscientific
are at the basis of the acrimony and controversy that contaminate the
field of psychotherapy.
Because of this, we cannot avoid the subject of metapsychiatry,

whether we like it or not. Only a theoretical understanding of
psychotherapy promises to help disengage us from some of the
unnecessary and acrimonious controversy in our field. The founder of

the subject was Freud, who defined metapsychology as the study of
the assumptions on which the system of psychoanalytic theory is
based. This is an unfinished study.
Although in a jocular vein Freud spoke of metapsychology as
the witch of psychoanalysis, he was insistent about the need for it as

a stable theoretical foundation for his empirical findings. Waelder
(1960) defined metapsychology as that level of abstract concepts
which lies between inductively constructed clinical theory and the

philosophical assumptions upon which the entire science is based.
The crucial scientific test of clinical theory is that of truth or validity;
for metapsychology, it is that of usefulness and internal consistency.
New empirical finds that do not fit into existing metapsychology

should lead to its revision. Such changes should be made, however, so
as not to disturb the internal consistency of the whole system. The
entire set of theories must not be treated as a rigid and fixed system;
on the other hand, it is equally sterile for a science to regard its
theories in an offhand or amorphous manner.
A more general epistemological question regarding the
relationship of metapsychology to reality and to scientific method has
been repeatedly raised, sometimes seriously and sometimes in a
pejorative manner. This kind of study I have defined as the subject of
metapsychiatry in previous publications (Chessick 1969, 1974b). We

ask the following questions in this discipline: (1) What is the position
of psychotherapy in the western philosophical tradition? (2) To what
extent can psychotherapy be said to be a science and to yield scientific

knowledge? (3) To what extent is psychotherapy a philosophy or an
art? Thus we might ask, Do generalizations based on the clinical data
of psychotherapy represent scientific knowledge? Where do such

propositions stand with respect to knowledge by intuition, knowledge
by philosophy, or knowledge obtained through the method of

science? The three kinds of propositions are often mixed together as
knowledge. All propositions are matters of opinion regarding truth
or reality, and there are no absolutely true propositions that are not at
the same time tautological.
Through a rigorous use of scientific method we approach
certainty with the greatest probability. However, many areas of study

simply do not lend themselves to scientific experimentation in the
rigid sense but must depend on the common accumulated historical
experience of mankind; for example, Slavery is always undesirable,

or Participatory democracy is the most advanced form of
government. I call such propositions philosophical, and they are
arrived at by the method of philosophy and suffer from a lesser
certainty. Propositions with the least certainty, such as insights or

religious intuitions, are arrived at by the method of intuition. These
are sometimes believed but never testable.
Generalizations from the clinical data of psychotherapy are on
the borderline between philosophical knowledge and scientific
knowledge because, although some experimental manipulation is

possible in restricted situations, no crucial experiments can be
devised that could lead to the definitive acceptance or rejection of a

system of generalizations from the clinical data of psychotherapy.
An outstanding example of metapsychiatry is provided by
Kohut (1971). How does one decide whether a specific form of

psychotherapy is primarily scientific or primarily inspirational? He
suggests asking three important questions: (1) Do we have a
systematical theoretical grasp of the processes involved in therapy?

(2) Can the treatment method be communicated to others, learned
and practiced without the presence of its originator? (3) Does the
treatment method remain successful after the death of its creator?
This latter question frequently separates out therapies that primarily

depend on the charisma of their originators.
One of the most important areas of metapsychiatry to have
generated much controversy and needless acrimony lies in the

understanding of what goes on in psychotherapy between the patient
and the therapist, often labeled psychotherapeutic interaction. The
basic trend in intensive psychotherapy since the time of Freud has

been to increasingly emphasize and understand the therapeutic
aspects of the relationship between the psychotherapist and the
patient, and we know that in the psychotherapy of the borderline

patient the psychic field and the deep inner attitude (Nacht 1963) of
the psychotherapist are absolutely crucial. We know that the
psychotherapist must have inborn talent, close supervision and
thorough knowledge of psychodynamics and therapeutic technique.
Furthermore, it is clear that an optimal psychic field must be
presented to the patient by the psychotherapist.
The traditional models of intensive psychotherapy have often
been based on the chess model first suggested by Freud (1913). It is
common knowledge that the opening and ending rules in a chess

game can be exhaustively analyzed but the middle game moves offer
innumerable creative possibilities; only general guidelines can be
taught, followed by careful analysis of the games of master players. In

chess each player influences the other continuously. Thus the same
player plays differently against different opponents, even though he
may have a persistent style of his own.
In a previous paper (Chessick 1971b) in which the parallel
between learning difficulties in chess and learning problems in

psychotherapy has been presented, I have emphasized the parallel
between chess blindness and the inherent difficulties in seeing
what the patient is trying to communicate in the myriad of material. I
have in a book (Chessick 1971c) presented a special theory of

psychotherapeutic interaction which attempts to minimize this loss of
understanding by better focus on and improved descriptions of the
mutual interaction between patient and therapist. Let me review this

briefly.
Just as the special theory of relativity holds only for certain
special situations (observers in uniform relative motion), I use the
phrase special theory (maintaining the analogy to physics) because
my theory also holds only for certain special situationsindividual
psychotherapy using the definitions, settings and techniques
generally accepted as constituting psychoanalytically oriented

psychotherapy (1969, 1974b).
A second analogy to the special theory of relativity is that my
theory can be reduced to Freudian psychodynamics for everyday

practical use if certain limitations are observed (similarly the special
theory of relativity can be reduced to Newtonian physics for practical
terrestrial use).
Whitaker and Malone (1953) developed a preliminary concept
to the special theory of psychotherapeutic interaction which they

labeled symbolic synchronization and complimentary articulation.
It rests on the belief that in all psychotherapy both participants have
both therapist and patient vectors within them. Their concept
received very little attention of a technical nature in the literature.

According to these authors, therapist vectors are defined as responses
to the needs of the immature part of the other person. Usually a

therapists responses are therapist-vector responses to the patient; at
times, however, the patient will respond with therapist-vector
responses to the relatively small (we hope), residual immature part of
the therapist. Patient vectors are demands for the expression of

feelings from the other person comparable to the demands of the
hungry child for a response from his parents. It is obvious that the
patient will get well only if the precondition is met that the therapists
patient vectors do not make excessive demands on the patients
therapist vectors.
Although Whitaker and Malone politely draw a contrast
between the gross pathological patient vectors of the immature
therapist, and the minimal, residual patient vectors in the mature

therapist, their main point is that successful psychotherapy requires
the therapist to bring along both his therapist and his patient vectors
and to engage in a total participation with the patient. The therapist

expands the frontier of his own emotional growth through the
therapy; if he refuses to participate totally in this fashion, the patient
experiences a rejection and therapy is a failure. Neither therapist nor

patient may even be aware of what is happening. Many experienced
psychotherapists seem to be able to confirm this by pointing out that
in each successful psychotherapy they experience some aspect of
further emotional growth, ego integration or maturationoften

called learning from the patient.
It is usually assumed that psychotherapy is in part an art

merely because of our ignorance about the field. This assumption
implies that as we gain more knowledgeor more precisely scientific
understandingof psychology and psychotherapy, the practice of

psychotherapy will become more and more scientific, thereby
approaching the ideal doctor-patient model in medicine.
I maintain that this generally held fundamental assumption is
completely wrong and it accounts for much of the confusion and

acrimony in our field as well as for unfair and invidious comparisons
with other more scientific branches of medicine. For this
assumption is based on a misconception about the nature of
knowledgenotice that we are again back to the subject of
metapsychiatry. This misconception, which has prevailed for
centuries, currently appears as a squabble between the proponents of

science and the proponents of the humanities, often defined as the
two cultures. The usual answer to this squabble is that with
understanding and time and patience the two cultures can become
one.
Some authors have challenged this popular answer directly.
For example, Levi (1963) bases the challenge on a study of the
philosophy of Kant. He argues that the disagreements and differences

between scientists and humanists are based upon ignoring the
distinction already found in Kant. Science focuses on facts and
basically relies upon a mechanistic formulation of the principles of

causation. The humanities are teleological, dramatic and emotional,
and they are oriented to human purposes in a manner that cannot be
allowed by the impersonality and objectivity of science. Levi writes,
The avowed and willing anthropocentrism of the humanities is far

removed from the neutral causation of science. Scientists and
humanists think differently and use different languages.
The language of science stresses true and false propositions,
error, causality, law, prediction, fact and equilibrium of systems. The

language of the humanistic imagination focuses upon destiny and
human purpose, fate and fortune, tragedy and illusion. It is certainly
possible to argue that depending on which critical faculty of the mind
imagination or understandingis being employed, an entirely
different map of what appears to be reality will emerge. One map will
be sober and factual, claiming to be the custodian of literal truth,
mechanistic and objective. The second will be mythical, teleological
and dramatic and will deal more with concepts of creativity, destiny
and human purpose. According to Levi the first will be based on
Kants synthetic a priori principles of the understanding, and the
second will be based on Kants concept of reproductive imagination

from the Critique of Pure Reason.
It is not possible to carry this argument from Kants
philosophy much further. Actually, in Levis interpretation there is

considerable debatable extrapolation from Kant. At any rate, as Levi
conceives of it, imagination is the human faculty from the active
functioning of which the humanities stem, whereas science is based
on the faculty that employs principles of cognitive understanding. The

basic point is that scientific understanding and humanistic
imagination are fundamentally different, utilize different languages,

provide different maps of reality, and are grounded on different
nuclear operations of the mind.
To construct objective, factual mechanistic chains of casual

explanations, as well as to construct heuristic, often dramatic and
anthropomorphic explanatory fictions, are both fundamental human
cognitive needs. The great physicist Bohr (Heisenberg 1971) similarly

distinguished among the languages of religion, science and art and
suggested, We ought to look upon these different forms as

complementary descriptions, which though they exclude one another,
are needed to convey the rich possibilities flowing from mans
relationship with the central order. Thus the language of the
imagination and the language of the understanding represent
different ways of looking at the same sensory manifold (of course,
sensory manifold is a term borrowed from Kant, although this
argument is no longer being presented in Kantian terminology).
The special theory of psychotherapeutic interaction takes into
account the different nuclear operations of the mind which may be

used in organizing the sensory manifold and makes it more
understandable how differences and arguments arise among
observers of the sensory manifold depending on what operations they

apply to it. We thus provide four roots of psychotherapeutic
interactions instead of the usual two. This is because maps of the
psychic fields interacting between the therapist and the patient must
be described in a bilingual fashion and the languages must not be

confused with each other. Each language selects a center for the
psychic field of the therapist and another analogous center for that of
the patient.
In the language of scientific understanding, the therapist may
be described in terms of his ego operations, countertransference

structure, therapist and patient vectors, and training in therapeutic
technique. In this language the patient may be described in terms of
his ego operations, a genetic-dynamic formulation, the structural

theory of Freud, transference, and patient and therapist vectors. Thus
a scientific understanding of the process of psychotherapy would

have to examine the study of mutual influencing throughout the
psychotherapy on both unconscious and conscious levels of the
psychic fields of the therapist and the patient, using the descriptive
terminology just outlined.
In the language of the humanistic imagination, which is much
more dramatic and emotional and is oriented to human purposes, the

two psychic fields would be described quite differently. Here we find
terminologies such as the I-Thou relationship, self-actualization, the
authentic life, the encounter, basic anxiety, the will to power, caring,
presence, the capacity for trust, life-style, career line, and even
Freuds famous statement that psychotherapy is a labor of love.
The quarrel between so-called opposing schools of

psychotherapy will arise in the contrast that naturally emerges when
the method of science or the method of the humanistic imagination is

applied to the same sense data. The two maps of reality and
descriptions of what is going on are not fundamentally opposed and
may be used successfully to complement each other, provided the

therapist is carefully aware when and why he is using each competing
map. If this is possible, then a greater understanding of patient
material and patient problems can be achieved and we can have
greater depth or conception of how to present the most effective

psychic field to the patient.
Obviously if this basic theoretical orientation is correct, a
corresponding education in the language of the humanistic
imagination will have to be provided for the psychotherapist so he

may move comfortably from one map of the sensory manifold to the
otherfrom the language of science to the language of creative
imagination. Kinzie and Jurgensen (1976) have made an effort to
apply this to the improvement of psychiatric education. The genius of
Freud was often expressed in his remarkable capacity to move back
and forth from the faculty of scientific investigation to the faculty of

creative and humanistic imagination. Because of his unusually wide
erudition and genius he often tended to switch back and forth
between these languages in order to present as immediate and

complete a description of the clinical phenomena as he could, and in
addition his contemporary readers had a significantly broader
background in the humanities than the average physician does today.

What subsequently happened, of course, is that the two languages
became confused in the minds of his less erudite followers and even
more in the minds of general readers, so that a number of
pseudoproblems arose, leading to various animosities that still exist.
An education strictly confined to the technique and practice of
scientific psychotherapy tends toward a stability and a withdrawal
from participation with the patient at a truly human level. On the
other hand, an education too heavily weighted in the humanities and

without the firm anchor of both scientific methodology and
dedication to the medical physicianly vocation causes a profound loss
of the scientific grounding and the objective observation aspects of

psychotherapy, with a consequent serious tendency to misunderstand
and even to go off the deep end and engage in bizarre and
unjustifiable procedures with patients. Neither of these extremes is
fair to the patient. Both of them are exploitation and they represent a
serious and inexcusable defect in the psychic field of the therapist.
Furthermore, it is clear that psychotherapists will be more
inclined to fail if all aspects of the psychotherapeutic interaction are
not taken into account or not understood. The best insurance against
failure in such cases would be the ability of the therapist to describe
the interaction in both languages and to visualize maps of both

psychic fields. Because of the limits of our knowledge or of our
capacities, we can sometimes express success or failure in
psychotherapy in one language, but not in the other. Thus a failure

that seems inexplicable from the point of view of scientific
understanding can sometimes be explained in the language of
humanistic imagination and vice versa.
The highest, most abstract and least empirically verifiable level
is of course that of metaphysics and metaphysical propositions
about Being, etc. Besides the powerful human tendency to push to the
limits and transgress the boundaries of what can be known by reason,

is there any other justification for our interest as psychotherapists in
metaphysics? I believe there is. Just as appeal to the philosophy of

science resolves some apparent paradoxes in the psychotherapy of
borderline patients, certain aspects of metaphysics explain some of
the clinical phenomena of borderline patients. This is forced upon us

whether we like it or not, because these patients often present their
complaints in the language level of metaphysics. They complain of
innumerable existential crises and difficulties in their sense of Being

and sense of aliveness; of alienation and isolation from the world,
from man and from God; of preoccupation with nothingness and so
on. It is wise not to just brush off such complaintsthey are very
meaningful to the patient and positively are not merely ways of
expressing depression.
Let us appeal, then, to the level of discourse of metaphysics in

an attempt to understand these complaints, which appear so strange
and vague at the level of empirical clinical examination.
The rock bottom of the entire theory of Kant is founded on our
inner awareness in time. All commentators on Kant agree with this. It

is from this inner awareness that he deduces his whole architectonic.
There is a parallel between Kant and Freud in that both agree that the
phenomena of the conscious are knowable and that from these
phenomena we have to deduce the existence of the stimuli from the

unconscious or noumena, which are basically unknowable. Thus the
id is knowable only through its derivatives in the ego. There are many
difficulties in this view, but for our purposes what is very important
to focus upon is the concept of the ego experience itself, which Kant
called our inner awareness of ourselves in time.
Winnicott (1968) writes, Good enough holding . . . facilitates
the formation of a psychosomatic partnership in the infant. This
contributes to the sense of real as opposed to unreal. Faulty
handling militates against the development of muscle tone, and that

which is called coordination, and against the capacity of the infant
to enjoy the experience of body functioning, and of Being. ... If the
environment behaves well, the infant has a chance to maintain a sense

of continuity of Being; perhaps this may go right back to the first
stirrings in the womb. When this exists the individual has a stability
that can be gained in no other way. What in psychodynamic terms is
this sense of Being and the continuity of Being which is called a
function of good enough holding in infants?
To answer, I (see also Chessick 1974b) am going to refer to
Federn (1952) as translated by Weiss. It is not always clear where

Federn is talking and where the translating, editing and introducing
by Weiss is presenting Federn or Weiss. At any rate these authors
present a difficult concept of the ego as a subjective experience. They

label this subjective experience the ego experience (Icherlebnis).
This phenomenon of the egos experience of itself cannot be

clearly explained. As long as the ego functions normally one may
ignore or be unaware of its functioning. As Federn says, normally
there is no more awareness of the ego than of the air one breathes;
only when respiration becomes burdensome is the lack of air
recognized. The subjective ego experience includes the feeling of
unity, continuity, contiguity and causality in the experiences of the
individual. In waking life the sensation of ones own ego is

omnipresent, and it undergoes continuous changes in quality and
intensity.
Federn sometimes distinguishes clearly and very carefully
between ego consciousness (Ichbewsstsein) and ego feeling
(Ichgefhl). He writes, Ego feeling is the sensation, constantly

present, of ones own personthe egos own perception of itself... we
can distinguish, often accurately, between ego feeling and ego
consciousness. Ego consciousness, in the pure state, remains only

when there is a deficiency in ego feeling. And the mere empty
knowledge of ones self is already a pathological state, known as
estrangement or depersonalization.
Ego consciousness represents an enduring feeling in our
knowledge that our ego is continuous and persistent despite

interruptions by sleep or unconsciousness. We feel that the processes
within us, even though they be interrupted by forgetting or
unconsciousness, have a persistent origin within us and that our body

and our psyche belong permanently to our ego. Ego consciousness is
an entity involving the continuity of a person in respect to time, space
and causality, and the sense of ego consciousness plays a central role
in the argument of Kants Critique of Pure Reason.
Ego feeling, however, is the totality of feeling which one has in

ones own living person. It is the residual experience which persists
after all the subtraction of all ideational contentsa state which, in
practice, occurs only for a very brief time. . .. Ego feeling, therefore, is
the simplest and yet the most comprehensive psychic state which is
produced in the personality by the fact of its own existence even in
the absence of internal stimuli.
To say the least, this is an extremely important and neglected
concept for both philosophers and psychotherapists. Federn explains
that ego feeling is quite different than new knowledge of ones self or
of consciousness of the ego at workit is primarily a feeling or
sensation normally taken for granted. This is parallel to Heideggers
(1954) explanation that The Being of beings is the most apparent;

and yet, we normally do not see itand if we do, only with difficulty.
Furthermore, both Heidegger and Federn would agree that . .. even
the clearest knowledge of ones ego is experienced as something
insufficient, uncomfortable, incomplete, and unsatisfying, even akin to
fear...
Freud (1917) used the same term ego feeling (Ichgefhl) in
Mourning and Melancholia, but he used it to mean something akin to
self-esteem, which is of course quite different. Also, notice that I am

not discussing or advocating Federns theory that schizophrenia
represents a deficiency of ego libido at allfor this is a different
discussion and a much more controversial concept. In general,
Federns thinking is ingenious and original but semantically confused,

as others have noticed.
The notion of ego feeling is extremely important to our
modern work with borderline patients. The capacity to develop
mature and secure relationships with other individuals and a strong
sense of inner sustainment is grounded on a healthy ego feeling. The

ego defect so frequently talked about in vague terms in describing the
borderline patient to a great extent is a defect in ego feeling. The

cause of this defect, or, as we may call it philosophically, disturbance
in the sense of Being, can be traced clinically in the borderline patient
to a lack of good enough holding in infancy. The result of such a falling
away from Being is that relationships become more uncherishing

and come to partake of the quality Buber calls I-It; the individual
often becomes immersed in an obsessive search for something he
intuitively knows is missing but cannot describe clearly in words.
Heidegger spent his life in an obsessive intellectual search for
Being, which he projects at least in part outside himself and which he
can never find. In this, he is like the typical borderline patient, who is
able to function very well in business and mundane matters, but, for
example, finds herself obsessed with the need for holding as in the
series of women described by Hollender (1970; Hollender et al. 1969,
1970) and a search for the magical sensation this produces. At an
extremely primitive level, these patients are searching for a

temporary sense of relatedness and ego feeling that is basically
missing in them and cannot be replaced with any kind of intellectual
or verbal exchange.
Those who have not experienced such problems clinically have
an extremely hard time understanding this set of concepts. The
patients have to teach us. As Heidegger (1953) put it, But an age
which regards as real only what goes fast and can be clutched with
both hands looks on questioning as remote from reality and as
something that does not pay, whose benefits cannot be numbered.

But the essential is not number; the essential is right time, i.e. the
right moment, and the right perseverance.
We have come a long distance in our discussion of the
borderline patient in this book, from Kraepelinean descriptive

psychiatry all the way to metaphysics. If a therapist has thoroughly
grasped all the concepts discussed in this book and has accumulated
twenty years of experience in working with borderline patients, he

may feel secure that he has acquired 10 percent of the preparation
necessary to practice psychotherapy. The other 90 percent can only
be acquired by a thorough intensive personal psychotherapy of the
psychotherapist.
Share this Book with your Friends!
References
Adler, G. 1973. Hospital treatment of patients with borderline

personality organization. International Journal of Psycho-
Analysis 49: 600-619.
Appelbaum, A. 1975. Transactions of the Topeka Psychoanalytic

Society. Bulletin of the Menninger Clinic 19: 384-390.
Arieti, S. 1974. Interpretation of Schizophrenia. New York: Basic

Books.
Balint, M. 1953. Primary Love and Psychoanalytic Technique. New

York: Liveright.
_____. 1968. The Basic Fault: Therapeutic Aspects of Regression.

London: Tavistock.
Basch, M. 1973. Psychoanalysis and theory formation. In Annual of

Psychoanalysis 1: 39-52.
_____. 1975. Toward a theory that encompasses depression: A revision

of existing casual hypotheses in psychoanalysis. In Depression
and Human Existence, ed. E. Anthony and T. Benedeck.
Boston: Little, Brown.
Blum, H. 1973. The concept of erotized transference. Journal of the

American Psychoanalytic Association 21: 61-76.
Bohr, N. 1934. Atomic Theory and the Description of Nature.

Cambridge, Mass.: Cambridge University Press.
Breuer, J., and Freud, S. 1895. Studies on Hysteria, Standard Edition. 2:

1-305 (1955).
Brodey, W. 1965. On the dynamics of narcissism. Externalization and
early ego development. Psychoanalytic Study of the Child 20:
165-193.
Cary, G. 1972. The borderline condition. Psychoanalytic Review 59: 33-

54.
Chessick, R. 1966. Office psychotherapy of borderline patients.

American Journal of Psychotherapy 20: 600-614.
_____. 1968. Crucial dilemma of the therapist in psychotherapy of

borderline patients. American Journal of Psychotherapy 22:
655-666.
_____. 1969. How Psychotherapy Heals. New York: Jason Aronson.
_____. 1971a. Amaurosis schacchistica. American Journal of

Psychotherapy. 25:309-311.
_____. 197 I b. The use of the couch in psychotherapy of borderline

patients. Archives of Psychiatry 25: 306-313.
_____. 1972a. Angiopastic retinopathy. Archives of Psychiatry 27: 241-

244.
_____. 1972b. Externational and existential anguish. Archives of

Psychiatry 27: 764-770.
_____. 1971c. Why Psychotherapists Fail. New York: Jason Aronson.
_____. 1973. Contributions to ego psychology from the treatment of

borderline patients. Medikon 2: 20-21.
_____. 1974a. Defective ego feeling and the quest for Being in the
borderline patient. International Journal of Psychoanalytic
Psychotherapy 3: 73-89.
_____. 1974b. The Technique and Practice of Intensive Psychotherapy.
New York: Jason Aronson.
_____. 1975. The borderline patient. In American Handbook of

Psychiatry, 2nd ed., ed. S. Arieti. New York: Basic Books.
_____. 1977a. Great Ideas in Psychotherapy New York: Jason Aronson.
_____. 1977b. Agonie: Diary of a Twentieth Century Man Ghent,

Belgium: European Press.
_____. 1977c. The coronary-prone personality. Medikon Internationale.

17-20.
Deutsch, H. 1965. Some forms of emotional disturbances and their

relationship to schizophrenia. In Neuroses and Character
Types. New York: International Universities Press.
DeWald, P. (1964). Psychotherapy. New York: Basic Books.
_____. 1972. The clinical assessment of structural change. Journal of the

Erikson, E. 1959. Identity and the Life Cycle. New York: International
Universities Press.
Evans, R. 1973. Jean Piaget. New York: Dutton.
Federn, P. 1952. Ego Psychology and the Psychoses. New York: Basic
Books.
Ferenczi, S. 1950. Stages in the development of the sense of reality. In

Selected Papers, vol. 1. New York: Basic Books.
Fairbairn, R. 1952. Object Relations Theory of the Personality. New

York: Basic Books.
Freud, A. 1965. Normality and Pathology in Childhood. New York:
International Universities Press.
_____. 1969. Difficulties in the path of psychoanalysis. In Writings of

Anna Freud, vol. 7. New York: International Universities
Press.
Freud, S. 1895. Project fora scientific psychology. Standard Edition 1:

283-387 (1966).
_____. 1909. Notes upon a case of obsessional neurosis. Standard

Edition 10: 153-318 (1957).
_____. 1913. On beginning the treatment. Standard Edition 12: 121-144

(1958).
_____. 1914a. On narcissism. Standard Edition 14: 67-104 (1957).
_____. 1914b. On the history of the psychoanalytic movement. Standard

Edition 14: 3-66 (1957).
_____. 1915. Observations on transference love. Standard Edition 12:

157- (1958).
_____. 1917. Mourning and melancholia. Standard Edition 14: 237-258

(1957).
_____. 1923. The ego and the id. Standard Edition 19: 3-68 (1961).
_____. 1924. Economic problems of masochism. Standard Edition 19:

157- (1961).
_____. 1927. Fetishism. Standard Edition 21: 149-158 (1961).
_____. 1938. An outline of psycho-analysis. Standard Edition 23: 141-

208 (1964).
_____. 1940. Splitting of the ego in the process of defense. Standard
Edition 23: 271-278 (1964).
Friedenberg, E. 1974. R. D. Laing. New York: Viking Press.
Friedman, H. 1975. Psychotherapy of borderline patients. American

Journal of Psychiatry 132: 1048-1051.
Friedman, L. 1975. Current psychoanalytic object relations theory and

its clinical implications. International Journal of
Frosch, J. 1971. Technique in regard to some specific ego defects in

the treatment of borderline patients. Psychiatry Quarterly 35:
216-220.
Gedo, H., and Goldberg, A. 1973. Models of the Mind. Chicago:

University of Chicago Press.
Giovacchini, P. 1965. Transference, incorporation, and synthesis.

International Journal of Psychoanalysis 46: 287-296.
_____. 1967a. The frozen introject. International Journal of

_____. 1967b. Frustration and externalization. Psychoanalytic Quarterly

36: 571-583.
_____. 1972. Tactics and Techniques in Psychoanalytic Therapy. New

York: Jason Aronson.
_____. 1973. Character disorders: with special reference to the

borderline state. International Journal of Psychoanalytic
_____. 1975a. Psychoanalysis of Character Disorders. New York: Jason
Aronson.
_____. 1975b. Tactics and Techniques in Psychoanalytic Therapy II:

Countertransference. New York: Jason Aronson.
Gitelson, M. 1962. On the curative factors in the first phase of analysis.

In Psychoanalysis. New York: International Universities Press.
_____. 1963. The problem of character neurosis. Journal of the Hillside

Hospital 12: 3-17.
_____. 1973. Psychoanalysis: Science and Profession. New York:

Goldberg, A. 1975. Narcissism and the readiness for psychotherapy

termination. Archives of Psychiatry 32: 695-704.
Greenacre, P. 1963. Problems of acting out in the transference

relationship. Journal of the American Academy of Childhood
Greenson, R. 1958. Screen defences, screen hunger, screen identity.

Journal of the American Psychoanalytic Association 6: 242-
262.
Grinker, R. 1955. Growth, inertia, shame. International Journal of

_____. 1975. Psychiatry in Broad Perspective. New York: International

Universities Press.
Grinker, R., Werble, B. and Drye, R. 1968. The Borderline Syndrome.

New York: Basic Books.
Gunderson, J., and Singer, N. 1975. Defining borderline patients.

American Journal of Psychiatry 132: 1-10.
Guntrip, H. 1968. Schizoid Phenomena, Object Relations and the Self.
New York: International Universities Press.
Hartmann, H. 1950. Comments on the psychoanalytic theory of the

ego. Psychoanalytic Study of the Child 5: 74-96.
_____. 1964. Essays on Ego Psychology. New York: International

Universities Press.
Hartmann, H., and Lowenstein, R. 1962. Notes on the superego.

Psychoanalytic Study of the Child 17: 42-81.
Havens, L. 1974. On the existential use of the self. American Journal of

Heidegger, M. 1953. An Introduction to Metaphysics. New York: Basic

Books.
_____. 1954. What is Called Thinking? New York: Harper and Row.
Heimann, P. (1966). Comment on Dr. Kernbergs paper. International

Journal of Psychoanalysis 56: 137-146.
Heisenberg, W. 1971. Physics and Beyond. New York: Harper and Row.
Hoch, P., and Catell, J. 1959. The diagnosis of pseudoneurotic

schizophrenia. Psychiatric Quarterly 33: 17-43.
Hoch, P., Catell, J., Strahl, M. and Pennes, H. 1962. The course and
outcome of pseudoneurotic schizophrenia. American Journal
of Psychiatry 119: 106-115.
Hoch, P., and Polatin, P. 1949. Pseudoneurotic forms of schizophrenia.

Psychiatric Quarterly 23: 248-276.
Hollender, M. 1970. The need or wish to be held. Archives of General
_____, Luborsky, L.,and Harvey, R. 1970. Correlates of the desire to be

held in women. Psychosomatic Research 14: 387-390.
_____, Luborsky, L., and Scaramella, T. 1969. Body contact and sexual
enticement. Archives of General Psychiatry 20: 188-191.
Holton, G. 1973. Thematic Origins of Scientific Thought. Cambridge,

Mass.: Harvard University Press.
Husserl, E. 1965. Phenomenology and the Crisis of Philosophy. New

York: Harper Torchbooks.
Jacobson, E. 1964. The Self and the Object World. New York:
Jaspers, K. 1963. General Psychopathology. Chicago: University of

Chicago Press.
Kahn, M. 1974. The Privacy of the Self. New York: International

Universities Press.
Kernberg, O. 1966. Structural derivatives of object relationships.

_____. 1967. Borderline personality organization. Journal of the

_____. 1968. The treatment of patients with borderline personality

organization. International Journal of the Psychoanalysis 49:
600-619.
_____. 1970a. Factors in the psychoanalytic treatment of narcissistic

personalities. Journal of the American Psychoanalytic
Association 18: 24-50.
_____. 1970b. A psychoanalytic classification of character pathology.
Journal of the American Psychoanalytic Association 18: 800-
822.
_____. 1971. Prognostic considerations regarding borderline

personality organization. Journal of the American
Psychoanalytic Association 19: 595-635.
_____. 1972a. Early ego integration and object relations. Annals of the
N. Y. Academy of Sciences 193: 233-247.
_____. 1972b. The Kleinian school. In Tactics and Techniques in

Psychoanalytic Therapy, ed. P. Giovacchini. New York: Jason
Aronson.
_____. 1973. Psychoanalytic object relations theory. In Annual of

Psychoanalysis, vol. 1, New York: Quadrangle Books.
_____. 1974a. Contrasting viewpoints regarding the nature and

psychoanalytic treatment of narcissistic personalities. Journal
of the American Psychoanalytic Association 22: 255-267.
_____. 1974b. Further contributions to the treatment of narcissistic

personalities. International Journal of Psychoanalysis 55: 215-
240.
_____. 1975a. Borderline Conditions and Pathological Narcissism. New

York: Jason Aronson.
_____. 1975b. Further contributions reply to discussion. International

Kinzie, J., and Jurgensen, R. 1976. Modes of experience and psychiatric

education. American Journal of Psychotherapy 30: 276-288.
Knight, R. 1953. Borderline states. In Drives, Affects, Behavior, ed. R.
Lowenstein. New York: International Universities Press.
_____. 1962a. Borderline states. In Psychoanalytic Psychiatry and

Psychology. New York: International Universities Press.
_____. 1962b. Management and psychotherapy of the borderline

schizophrenic patient. In Psychoanalytic Psychiatry and
Psychology. New York: International Universities Press.
Kohut, H. 1959. Introspection, empathy and psychoanalysis. Journal of

the American Psychoanalytic Association 7: 459-483.
_____. 1966. The forms and transformations of narcissism. Journal of

the American Psychoanalytic Association 12: 243-272.
_____. 1971. The Analysis of the Self. New York: International

Universities Press.
_____. 1972. Thoughts on narcissism and narcissistic rage. The

Kuhn, T. 1972. The Structure of Scientific Revolutions. Chicago:

University of Chicago Press.
Langer, S. 1942. Philosophy in a New Key. New York: Mentor Books.
Lazare, A. 1971. The hysterical character in psychoanalytic theory.

Archives of General Psychiatry 25: 131-137.
Leuba, J. 1949. Introduction a Letude Clinique du Narcissisme. Revue

Francaise de Psychoanalyse T. XIII, 4: 456.
Levi, A. 1963. The Six Great Humanistic Essays of John Stuart Mill. New
York: Washington Square Press.
Little, M. 1966. Transference in borderline states. International
Loewenstein, R. 1972. Ego autonomy and psychoanalytic technique.

Psychoanalytic Quarterly 41: 1-23.
Mack, J. 1975. Borderline States in Psychiatry. New York: Grune and

Stratton.
Mahler, W. 1952. Child psychoses and schizophrenia. Psychoanalytic

Study of the Child 7: 286-305.
_____. 1963. Thoughts about development and individuation.

_____. 1971. A study of the separation-individuation process.

_____. 1974. Symbiosis and individuation. Psychoanalytic Study of the

Child 29: 89-106.
_____. 1975. On Human Symbiosis and the Vicissitudes of Individuation.

New York: International Universities Press.
Mahler, M., et al., 1975. Psychological Birth of the Human Infant. New
York: Basic Books.
Mahler, M., and Gosslinger, B. 1955. On symbiotic child psychoses.

Mahler, M., and LaPerrier, K. 1965. Mother-child interaction during

separation-individuation. Psychoanalytic Quarterly 34: 483-
498.
Maltsberger, J., and Buie, D. 1974. Countertransference hate in the

treatment of suicidal patients. Archives of Psychiatry 30: 625-
633.
Masterson, J. 1972. The Treatment of the Borderline Adolescent. New
York: Wiley.
Masterson, J., and Rinsley, D. 1975. The borderline syndrome.

Meissner, W. 1974. Correlative aspects of introjective and projective

mechanisms. American Journal of Psychiatry 131: 176-180.
Meyer, L. 1974. Concerning the sciences, the arts, and the humanities.
Critical Inquiry 1: 163.
Modell, A. 1963. Primitive object relationships and the predisposition

to schizophrenia. International Journal of Psychoanalysis 44:
282-292.
_____. 1968. Object Love and Reality. New York: International

Universities Press.
_____. 1975a. The ego and the id: Fifty years later. International Journal
of Psychoanalysis 56: 57-68.
_____. 1975b. A narcissistic defense against affects and the illusion of

self-sufficiency. International Journal of Psychoanalysis 56:
137-146.
Murphy, W. 1973. Narcissistic problems in patients and therapists.

International Journal of Psychoanalytic Psychotherapy 2: 113-
124.
Murray, J. 1964. Narcissism and the ego ideal. Journal of the American
Nacht, S. 1963. The non-verbal relationship in psychoanalytic

treatment. International Journal of Psychoanalysis 44: 328-
333.
_____. 1969. Reflections on the evolution of psychoanalytic knowledge.
Psychoanalysis 50: 597.
Nagera, H. 1967. The concepts of structure and structuralization.

Nef, J. 1967. The United States and Civilization. Chicago: University of

Chicago Press.
Odier, C. 1956. Anxiety and Magic Thinking. New York: International

Universities Press.
Ornstein, P. 1973. On narcissism. Bulletin of the Philadelphia

Association for Psychoanalysis 23: 327-329.
_____. 1974a. Discussion of the paper by Otto Kernberg. International

_____. 1974b. On narcissism: Beyond the introduction. Annual of

Piaget, J., and Inhelder, B. 1966. The Psychology of the Child. New York:
Basic Books.
Pfeiffer, E. 1974. Borderline states. Diseases of Nervous System (May)

212-219.
Pruyser, P. 1974. What splits in splitting? A scrutiny of the concept

of splitting in psychoanalysis and psychiatry. Bulletin of the
Menninger Clinic 39: 1-46.
Pulaski, M. 1971. Understanding Piaget. New York: Harper and Row.
Ranged, L. 1955. Panel Report in the Journal of the American

Rapaport, D. 1944. The scientific methodology of psychoanalysis.
Collected Papers, ed. M. Gill. New York: Basic Books (1967).
_____. 1951. The autonomy of the ego. Bulletin of the Menninger Clinic
15: 113-123.
Riesman, D., Denney, R. and Glazer, N. 1955. The Lonely Crowd. New
York: Doubleday.
Russell, B. 1967. Autobiography, vol. 1. Boston: Little, Brown.
Sadow, L. 1969. Ego axis in psychopathology. Archives of General

Saul, L. 1958. The Technique and Practice of Psychoanalysis.

Philadelphia: Lippincott.
_____. 1970. Inner sustainment. Psychoanalytic Quarterly 39: 215-222.
Schafer, R. 1968. Aspects of Internalization. New York: International

Universities Press.
Schmideberg, M. (1955). The borderline patient. In American

Handbook of Psychiatry, vol. 1, ed. S. Arieti. New York: Basic
Books.
Spitz, R. (1965). The First Year of Life. New York: International

Universities Press.
Stern, A. 1938. Psychoanalytic investigation of and therapy in a

borderline group of neuroses. Psychoanalytic Quarterly 7:
467-489.
_____. 1945. A psychoanalytic therapy in the borderline neuroses.

Psychoanalytic Quarterly 14: 190-198.
_____. 1948. Transference in the borderline neuroses. Psychoanalytic
Quarterly 17: 527-528.
Stone, L. 1961. The Psychoanalytic Situation. New York: International

Universities Press.
Strupp, H. 1973. Psychotherapy, Clinical Research, and Theoretical

Issues. New York: Jason Aronson.
_____. 1975. Psychoanalysis focal psychotherapy and the nature of

the therapeutic influence. Archives of Psychiatry 32: 127-135.
Ticho, E. 1970. Differences between psychoanalysis and

psychotherapy. Bulletin of the Menninger Clinic 34: 128-138.
Toffler, A. 1970. Future Shock. New York: Random House.
Toulmin, S. 1960. Philosophy of Science. New York: Harper

Torchbooks.
Tower, L. 1956. Countertransference. Journal of the American

Waelder, R. 1960. Basic Theory of Psychoanalysis. New York:

Wangh, M. 1974. Concluding remarks on technique and prognosis in

the treatment of narcissism. Journal of the American
Weiner, N. 1973. The ego ideal and melancholia. Bulletin of the

Department of Mental Health Sciences, Hahnemann Medical
College and Hospital 2:4-17.
Whitaker, C., and Malone, T. 1953. The Roots of Psychotherapy.

Philadelphia: Blakiston.
Wilie, D. 1972. Negative countertransference, therapist
discouragement. International Journal of Psychoanalytic
Winnicott, D. 1951. Transitional object and transitional phenomena.

Collected Papers. New York: Basic Books.
_____. 1958. Collected Papers. New York: Basic Books.
_____. 1965. The Maturational Processes and the Facilitating

Environment. New York: International Universities Press.
_____. 1968. The Family and Individual Development. London:

Tavistock.
Wolberg, A. 1973. The Borderline Patient. New York: International

Medical Book Corporation.
Zavitzianos, G. 1974. Transference, therapeutic alliance and the

principle of multiple function of the ego. Journal of the
Philadelphia Association for Psychoanalysis 1: 298-313.
Zetzel, E. 1971. A developmental approach to the borderline patient.

American Journal of Psychiatry 128: 867-871.
Zetzel, E., and Meissner, W. 1973. Basic Concepts of Psychoanalytic

Psychiatry. New York: Basic Books.
Zucker, O. 1963. Psychoanalytic assessment of ego weakness.

American Journal of Psychotherapy 17: 275-285.

Intensive Psychotherapy of The Borderline Patient

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Intensive Psychotherapy of The Borderline Patient

Uploaded by

Copyright:

Available Formats

INTENSIVE

RICHARD D. CHESSICK, M.D.

e-Book 2017 International Psychotherapy Institute

All Rights Reserved

This e-book contains material protected under International and

Created in the United States of America

Part I HISTORICAL AND CLINICAL DESCRIPTION

3 State of the Ego

4 Typical Borderline Complaints

7 Disputes and Disagreements

8 Developmental Failure in the Borderline Patient

Part III DEVELOPMENTAL PATHOLOGY

9 Early Ego Development and Projective Identification

10 Internalized Object Relations

11 Unresolved Metapsychological Problems: My Views

Part IV PSYCHOTHERAPYGENERAL APPROACH

16 The Psychic Field of the Psychotherapist

Part V PSYCHOTHERAPYSPECIAL PROBLEMS

18 Transference in the Narcissistic Personality Disorder

19 Ambience of the Treatment

20 Rage and Externalization

22 Helpful Clinical Suggestions

23 Improvement and Repair

24 Concepts of Cure in Intensive Psychotherapy

26 Metapsychiatry and Beyond

Introduction to the Concept

The concept of the borderline patient is poorly understood and

psychiatrists wish to entirely discard the termborderline of what?

and treatment of the borderline patient. It is written from the point of

view of the psychotherapist; the clinician who treats patients day in

patient to the office practice of psychoanalytically informed

The best overall review article (Gunderson and Singer 1975),

which attempts to identify areas of agreement in the literature,

reviews eighty-seven referencesmany of which on careful

number of features most authors believe seem to characterize

borderline patients. These features are

1. The presence of intense affect, usually of a strongly hostile or

2. A history of impulsive behavior of all sorts

4. Brief psychotic experiences, which are likely to have a paranoid

5.Bizarre, dereistic, illogical or primitive responses on

The borderline patient represents a frequently encountered

type of patient, posing special problems for the psychotherapist and

al., 1968) using meticulous methodology, to try to better delineate the

syndrome a disease, an arrest of development, an emotional

biological, medical, psychological or social. The borderline, like health

and illness, is a system in process occurring in time: developing,

I think that we have to stay with this rather profound

conclusion also in the light of current knowledge on the subject.

state of our knowledge on the subject does not permit a resolution of

The best one can do at this time is to delineate the areas of

differences. Summoning our patience, we then must look further at

the clinical material in the true tradition of psychoanalytically

oriented psychotherapy and hope eventually, by further and further

disagreements. As Freud wrote (1914), I was not subject to influence

themselves begin to speak.

The reader will discover fundamental and unresolvable

differences of opinion as to the foundations of psychic development,

and although we can discard a number of theories of basic

common sense or reading like science fiction, there still is a

fundamental disagreement in the literature from at least two basic

entirely different theories of the development and treatment of the

making it clinically respectable goes to Stern. In three papers (Stern

and psychodynamic picture in broad outline and discussed special